Evidence-Based Psychodynamic Psychotherapies

Peter Fonagy, PhD1 This piece essays to cover the same ground as the preceding chapter by Westen and his collaborators, but with an even larger reference list that more comprehensively covers the wider psychotherapy research world outside the United States, with Fonagys own arena of work centrally in the United Kingdom. Fonagys meta-analysis of evidence-based outcome research studies is centered more on (research) efficacy rather than (naturalistic) effectiveness studies, and progresses systematically through the assessment of diagnosisspecific disorders, starting with major depressions, and going through varieties of anxiety disorders, eating disorders, substance abuse disorders, and personality disorders. In proceeding in this manner, Fonagy uses the traditional DSM conceptions of the categorization of discrete mental illnesses which this entire volume is an effort to transcend, both diagnostically and therapy assessment-wise, and is of course in contrast with the earlier papers in this section by Blatt et al., and by Shedler and Westen, which propose more complex psychoanalytically based understandings of mental functioning and the therapeutic process as the basis for diagnostic and treatment assessments. Fonagy shows, however, that even when using DSM symptom clusters there is considerable evidence of efficacy for psychodynamically-based therapies for a variety of disorders and shares with Westen et al., a meta-analytic portrayal of how much can be learned about treatment results from within the more limited DSM framework.

INTRODUCTION
This review is based on an exhaustive review of the psychotherapy outcome literature, undertaken originally at the instigation of the UK Department of Health by Roth and Fonagy (Department of Health, 1995). We have recently updated this review (Fonagy, Target, Cottrell, Phillips, & Kurtz, 2002; Roth & Fonagy, in press) and extended it to identify all studies of psychoanalytic psychotherapy. The usual methods for identifying studies were employed (Fonagy, Target et al., 2002; Roth & Fonagy, in press)]. The key questions that should be asked of this literature given the current state of research in this area (also see Westen, Morrison, & ThompsonBrenner, 2004) are: (1) are there any disorders for which short-term psychodynamic psychotherapy (STPP) can be considered evidence-based; (2) are there any disorders for which STPP is uniquely effective as either the only evidence based treatment or as a treatment that is more effective than alternatives, (3) is there any evidence base for long-term psychodynamic psychotherapy (LTPP) either in terms of achieving effects not normally associated with short-term treatment, or addressing problems which have not been addressed by STPP? In this context short-term therapy is conceived of as a treatment of around 20 sessions carried out usually once weekly. From the standpoint of psychodynamic psychotherapy, the database of research studies has significant limitations. Westen and colleagues (Westen et al., 2004) recently offered a powerful critique of the research methods used to assign the status of empirically supported or unsupported therapies. Research that is considered empirically supported tends to have three characteristics: (1) studies address a single disorder (usually Axis I) with diagnostic assessments to ensure
1

Freud Memorial Professor of Psychoanalysis, University College, London, England

134

homogeneity of samples, (2) treatments are manualised and are of brief and fixed duration to ensure the integrity of the experimental manipulation, (3) outcome assessments focus on the symptom(s) that represent the declared priority of the study (and often the intervention). The underlying aim is the maximisation of internal validity by random assignment, controlling confounding variables, and standardising procedures. Westen et al. (2004) identify four poorly supported assumptions underpinning the application of randomised controlled trial (RCT) methodology to psychotherapy research: (1) that psychopathology is so malleable that a brief intervention is likely to change it permanently, (2) that most patients can be treated for a single disorder or problem, (3) that psychiatric disorders can be treated with psychosocial interventions without regard to personality factors that are less likely to change with brief treatments, (4) that experimental methods provide a helpful gold standard for evaluating these packages. In reality, most forms of psychopathology encountered in specialist centres are treatment resistant (Kopta, Howard, Lowry, & Beutler, 1994) and co morbid with other disorders (Kessler, Stang, Wittchen, Stein, & Walters, 1999) which need to be tackled in the broader context of the patients personality structure (e.g.Thompson-Brenner, Glass, & Westen, 2003) and experimental methods need to be supplemented by correlational analysis to ascertain the effective components of treatment (Ablon & Jones, 2002).

METHODOLOGICAL

CONSIDERATIONS

Evidence for Evidence-Based Practice This review is based on what funders and researchers currently regard as appropriate evidence (Clarke & Oxman, 1999). The criteria which are used to determine what counts as evidence based practice must themselves be empirically tested. Their specificity (the likelihood of falsely identifying a treatment as effective) and sensitivity (the chance of misclassifying an effective treatment as ineffective) should be established against a variety of other public health criteria. The same empirical standards should be applied to these criteria as would be expected in association with other clinical decision making tasks. Face validity, which is what we currently have, is clearly insufficient. Treatments designated as evidence based by some criteria must be distinguishable from treatments that do not meet these criteria on several concurrent independent but relevant indicators such as theoretical coherence, public health impact, user/consumer acceptability, etc. The Absence of Evidence or Evidence for Ineffectiveness Current categorization in evidence-based psychotherapies conflates two radically different groups of treatments: those that have been adequately tested and found ineffective for a patient group, and those that have not been tested at all. It is important to make this distinction, since the reason that a treatment has not been subjected to empirical scrutiny may have little to do with its likely effectiveness. It may have far more to do with the intellectual culture within which researchers operate, the availability of treatment manuals, and peer or third party payer perceptions of the value of the treatment (which can be critical for both research funding and publication). The British psychodynamically oriented psychiatrist, Jeremy Holmes (2002), has eloquently argued in the British Medical Journal that the absence of evidence for psychoanalytic treatment should not be confused with evidence of ineffectiveness. In particular, his concern was that cognitive therapy would be adopted by default because of its research and marketing strategy rather than its intrinsic superiority. He argued that: (1) the foundations of cognitive therapy were less secure than often believed; (2) the impact of Cognitive Behavioural Therapy (CBT) on long term course of

135

psychiatric illness was not well demonstrated; (3) in one real life trial at least the CBT arm had to be discontinued because of poor compliance from a problematic group of patients who nevertheless accepted and benefited from couples therapy (Leff et al., 2000); (4) the effect size of CBT is exaggerated by comparisons with waiting list controls and (5) the emergence of a postcognitive behaviour therapy approach (e.g. Teasdale et al., 2000; Young, 1999) that leans increasingly on psychodynamic ideas. Whilst I am entirely in sympathy with Jeremy Holmes perspective, even if my work with Tony Roth was part of the target of his criticism, it is only fair to expose the shortcomings of his communication. Nick Tarrier (2002), in a commentary on Holmes piece, writes with passion: Holmes relies on the specious old adage that absence of evidence is not evidence of absence [of effectiveness]. [] I would have more enthusiasm for this argument if traditional psychotherapy were new. It has been around for 100 years or so. The argument, therefore, becomes a little less compelling when psychotherapys late arrival at the table of science has been triggered by a threat to pull the plug on public funding because of the absence of evidence (p. 292). Tom Sensky and Jan Scott (Sensky & Scott, 2002) were similarly distressed, both by Holmes selective review of evidence and his allegations that some cognitive therapists are starting to question aspects of their discipline. The message from the CBT camp is this: If psychodynamic clinicians are going to address the issue of evidence based practice, they will have to do more than gripe and join in the general endeavour to acquire data. Of course, psychodynamic clinicians are at a disadvantage and not simply because they are late starters (after all, many new treatments find a place at the table of evidence based practice). There are profound incompatibilities between psychoanalysis and modern natural science. Paul Whittle (2000) has drawn attention to the fundamental incompatibility of an approach that aims to fill in gaps in self-narrative with cognitive psychologys commitment to minimal elaboration of observations, a kind of Wittgensteinian cognitive asceticism. The making of meaning around a life narrative is fundamental to human nature. It is therefore inconceivable that psychoanalysis (or any process very much like it) will ever not be part of the range of approaches that people with mental health problems desire. However in this context, success is measured as eloquence (or meaningfulness) which is not reducible to either symptom or suffering. Moreover, psychoanalytic explanations invoke personal history, but behaviour genetics has brought environmental accounts into disrepute. While cognitive behavioural therapy also has environmentalist social learning theory at its foundations, it has been more effective in moving away from a nave environmentalist position. To make matters worse, within psychoanalysis there has been a tradition of regarding the uninitiated with contempt, scaring off most open-minded researchers. The engagement of psychoanalytic clinicians in research programs is a desirable goal. Psychoanalysts are more than ever before, but not yet fully, committed to systematically collecting data with the potential to challenge and contradict as well as to confirm cherished ideas. The danger that must be avoided at all costs is that research is embraced selectively only when it confirms previously held views. This may be a worse outcome than the wholesale rejection of the entire enterprise of seeking evidence since it immunizes against being affected by findings at the same time as creating an illusion of participation in the virtuous cycle of exploring, testing, modifying and re-exploring ideas. The Scope of the Task Most United Kingdom evidence-based treatment reviews have been uniquely based on RCTs. RCTs in psychosocial treatments are often regarded as inadequate because of their low

136

external validity or generalizability (Anon, 1992). In brief, it is claimed that they are not relevant to clinical practicea hotly debated issue in the field of psychotherapy (Hoagwood, Hibbs, Brent, & Jensen, 1995) and psychiatric research (Olfson, 1999). There are a number of well publicized reasons, which I have not the space to go into here, as to why randomized trials in many areas of health care may have low external validity: (1) the unrepresentativeness of health care professionals participating, (2) the unrepresentativeness of participants screened for inclusion (in order to maximize homogeneity), (3) the possible use of atypical treatments designed for a single disorder and (4) limiting the measurement of outcome to the symptom that is the focus of the study and is easily measurable but far from the only relevant dimension (Fonagy, 1999a). RCTs cover only a limited number of treatments, and many treatments remain unevaluated in relation to many conditions. Because there are at least 200 disorders of child and adult mental health (by DMS criteria) and many hundreds of different forms of intervention, most of which have many components and many characteristic patterns of delivery, it is inconceivable that a matrix of types of therapy by types of disorder could ever be populated by appropriate studies (Goldfried & Wolfe, 1996). This is no trivial issue. Studies that attempt to identify which component of a treatment program is essential to its success frequently find that apparently most of the layers of the onion can be removed and the effect be still there. Because outcome studies rarely help identify either the effective elements of treatments or the process of change that leads to improvement, many traditional influential supporters of outcome investigations are calling for fewer rather than more outcome studies. Alan Kazdin (1998) for example, recommended a socalled dismantling strategy that one by one removes potential components of change until the genuinely effective component is identified. Some have suggested that meta-analyses might offer a direct solution to this problem (e.g. Borkovec & Ruscio, 2001) but this is by no means a straightforward task. A recent meta-analysis by Wilson McDermut and colleagues (McDermut, Miller, & Brown, 2001) identified group therapy as efficacious for depression with average ES of 1.06. However, the group treatments involved teaching a wide range of different strategies in different studies (self-control techniques, problem-solving skills, relaxation skills, disputation of negative thinking). So was teaching a skill truly addressing a depression related deficit? Matching patients to treatments that emphasized the specific deficits they presented with did not increase the effect size, and even attention-control groups resulted in a reduction of symptoms. Yalom (1995) outlined 11 therapeutic factors in group therapy but none of the studies reviewed by McDermut et al. discussed any one of these. We simply do not know what aspect of group treatments for depression make them so effective. Beyond these fairly well publicized issues, the question arises of whether manualized treatments or treatment packages are the appropriate level of analysis in our search for effective interventions. For example, a study by Olfson and colleagues (Olfson, Mechanic, Boyer, & Hansell, 1998) followed up schizophrenic patients discharged from hospital, and found that patients who had contact with the outpatient clinician prior to discharge were better off in terms of symptom reduction than those who had no communication with outpatient staff. Similar, apparently minor, process parameters of care may be far more important in determining outcome than entire treatment packages. It is hard to imagine that a sufficient number of RCTs could ever be performed to assess all such, potentially key, parameters of care. Evidence based practice needs to look beyond the current database and look at practice-based evidence in order to comprehensively establish evidence based practice.

137

The Ideal Outcome Research Program This was proposed by Alan Kazdin a number of years ago, but as it would require us to rethink our entire approach to outcome studies and evidence-based practice it is unlikely ever to be implemented. Basically he suggests that (1) treatment research should begin with the identification of key dysfunctions associated with a disorder and the empirical demonstration of these dysfunctions in a sizeable proportion of the clinical group. (2) Further, a conceptual link must be established between a proposed treatment method and the dysfunctional mechanism hypothesized to underpin the disorder. (3) Only when this has been done can manualization commence, followed by (4) the collection of the hierarchy of evidence that forms the body of systematic reviews. (5) Process-outcome studies can then be implemented to establish key treatment components and necessary treatment length. Experimental studies of hypothesized processes and mechanisms need to confirm the correlational findings from process-outcome investigations. Finally, the boundary conditions for the treatment need establishing, in terms of patient and environmental characteristics which promote or undermine the effectiveness of the therapy. You will notice that this is a radically different approach to the one normally undertaken where the starting point is the evaluation of a designated treatment. Currently the identification of key psychological processes are at best post hoc. No wonder there are too many different treatment modalities. No wonder that we know so little about why any of them work. Reversing this process would be a remarkable achievement of scientific enterprise. Pragmatic Trials The answer to the controversy between efficacy and effectiveness studies of psychotherapy may lie in so called pragmatic or real-world trials. These minimal effort trials require experimentation in addition to ongoing outcome measurement. The experimental component of pragmatic trials includes randomization to alternative methods of care. Importantly, non-specific aspects of care are controlled under these circumstances, yet questions of direct relevance to the clinicians may be asked and answered. Patients who participate naturally reflect the clinical population and exclusion criteria are kept to a minimum. Comparison treatments are with routine practice which usually involves combination treatments and treatments titrated according to the clients response. The pragmatic trial imposes minimal constraints on management. The only major sacrifice to internal validity is the loss of blindness in assessment. Blindness, which is likely to be imperfect in psychosocial treatments in any case, may offer little advantage as regards objectivity of outcome assessment. Double blindness imposes unrealistic restrictions even upon routine pharmacological care, and deviations from normal practice threaten the validity and generalizability of any cost data used in the estimation of cost effectiveness. Concealment of allocation (the prevention of foreknowledge about the group to which the patient will be allocated if recruited) that is an important source of selection bias, is readily achievable in this context. The unique feature of such trials lies in the relevance of the questions that clinicians may ask of their routine practice. Ideally, clinical equipoise (genuine uncertainty concerning outcome) should drive the search for evidence. In evidence-based practice, clinical curiosity is sadly, rarely the motivator. Pragmatic trials could be a key additional line of information for evidence based practice. In combination with more rigorous RCTs (particularly relevant to new treatments) and the judicious use of observational data, they will provide evidence of sufficient richness to significantly advance standards of mental health care. The establishment and support of a profession-wide methodology for pragmatic trials should be considered an important additional task of evidence based practice psychotherapy initiatives.

138

Clinical Guidelines Closing the gap between practice and evidence brings us to another quantum leap in the sophistication with which evidence for psychotherapeutic clinical psychological services is considered. It is increasingly recognised that evidence does not speak for itself and to be usefully applied it needs to be reviewed and integrated by a group of unbiased experts, including individuals whose expertise is as users and carers. Clinical guidelines that integrate evidence and front-line experience drawn up by multi-disciplinary panels, which as a collection of individuals have full awareness of the limitations of everyday clinical practice, is a key step that has all too often been omitted in the past when the sole experience applied to the interpretation of evidence was one of management. Research, with its focus on selected patient populations, cannot, of course, tell clinicians what to do with specific individuals. Clinicians have to ask the research database specific questions with an individual patient in mind. How to pose this massive accumulation of such questions and, even more challengingly, how to obtain meaningful answers? These are far more complex skills than that of generating a systematic review. Many hope that clinical guidelines can and will perform the role of translation of research into practice increasingly well. The controversy that surrounds this issue is beyond the scope of this paper. It is perhaps sufficient for us to say that we cannot see guidelines, however sophisticated, ever substituting for clinical skill and experience any more than the Highway Code can substitute for skilled driving. Future research should perhaps look at the skill with which clinicians implement particular treatments and the relationship of that to patient outcome. In addressing the failure of translation of guidelines into clinical behaviour (Chilvers, Harrison, Sipos, & Barley, 2002; Higgitt & Fonagy, 2002), it is useful to differentiate between diffusion, dissemination and implementation (Palmer & Fenner, 1999). These are interrelated and increasingly active phases of a process. Publication in a journal article (diffusion) is a passive form of communication, haphazard, untargeted and uncontrolled (seemingly insufficient to achieve much in the way of change in clinical practice). The development of practice guidelines, overviews etc. are more active and targeted to an intended audience (dissemination). Implementation is yet more active, with sanctions and incentives, monitoring and adjustment to local needs. The methods for translating guidelines into practice range from the use of written materials, through educational efforts, via product champions, financial incentives and patient mediated interventions through to reminder systems. Notwithstanding problems of the currency of guidance, there is a very real question about the extent to which guidance is utilized. At a recent Australian meeting to review the fate of 14 guidelines, none were found to have fared well. The shorter they were, the more likely they were to have had a noticeable impact. Successful implementation was most likely if it was initiated at a local level. The Brain and the Mind The question is not is psychotherapy still relevant in the days of Prozac? but how can we make it really count against a background increasingly fragmented social support systems, and how can we use it efficiently where current medication is not enough? Drugs and psychotherapy both work to the extent that they do because they both impact on the functioning of the brain. How else could it be, since the brain is the organ of the mind? And diseases of the mind are unequivocally diseases of the brain. The outcome of psychotherapy therefore should be as easy (or easier) to measure in terms of brain function as in terms of behaviour or subjective reports.

139

But the brain is not the final frontier of our knowledge about the mind. Emotions may be crudely changed by drugs but without giving any meaning to the experience of mental disorder, or to the drug-induced change. Psychotherapy is the crystallization of the principle of psychological causation that mental disorder in many instances may be most usefully seen in psychological terms, as the product of specific beliefs, desires and emotions. Without understanding mental disorder psychologically it would be impossible to understand the self-evident social pathways to mental disorder: Poverty, unemployment, incest, homelessness, spiritual despair at the violence and heartlessness of abusive parenting, the almost limitless methods that people can find of inflicting human misery; all these ignominies, influence people's expectations about others, the trust that they may be capable of feeling, their anger about their treatment, the complex ways we all find to learn to live in the social context that the fortunes or misfortunes of our birth have presented for us. Turning away from therapy, could mean shutting our ears to such anguish. Psychotherapists blew the whistle on the prevalence and long term impact of child maltreatment. It is the feelings, wishes, beliefs, thoughts and desires in the wake of despair that psychotherapy must be retained to address. Severe social disadvantage increases the risk of mental disorder by orders of magnitude. That this is a psychological rather than a merely physical process is borne out by the predictive power of relative rather than absolute destitution. The further down the social pile, the greater the misery, regardless of the actual material wealth of the individual. It is felt disadvantage that is psychologically toxic. Of course, in this, and every other, case it can be argued that experience of deprivation is a brain state, but the logical conclusion from that argument was foreshadowed in Huxley's Brave New World. No-one would seriously suggest that the most appropriate ethical way of addressing the gigantic issues raised by social disadvantage might be the suppression by medication of the misery of social exclusion? But the native land of psychotherapy, the mental world of beliefs, desires and emotions -central for a while in 20th Century psychiatry -- is a fragile creature. Maintaining the causal significance of meaning runs the gauntlet of a powerful human need for concreteness and simplicity, which the physics and biology of brain research represent. Only a psychological Luddite would ignore the immense benefit we have drawn from brain research. But we would be equally deluded were we to deny its cost. Brain research has impacted our culture. The answers it has provided, when translated into exuberant media sound bites, have undermined the ethos of seeking for psychological meaning in the way therapists work with patients. The natural human desire to create a narrative, a story, around ones experience has given way under economic as well as political pressures to a profound disrespect for the mental, born less of disillusionment with the efficacy of psychotherapy than of the reductionism of certain biological psychiatrists. The United States over the last 15 years has witnessed a devastation of concern with the psyche. Many HMOs (protectors of insurance company profit margins) will fund either no psychotherapy or at most a half a dozen sessions. Not surprisingly, and coincidental with the biological revolution in psychiatry, spending on mental health care in the US has decreased in real terms by a factor of 50%. Psychotherapy is essential to whole-person mental health care. We cannot afford to abandon it, if we are to offer meaningful and respectful care to those in distress. My concern here is with the risk of irreparable, even if unintended, damage to a perspective which enshrines in psychiatry what is unique about our species - that we recognise in each other the presence of a mind, the presence of emotions, wishes and beliefs as the motivators of behaviour, adaptive or maladaptive. Turning away from psychological therapy, from the truth of the importance of mind,

140

risks apocalyptic cultural and social changes, inevitable if we mock personal experience and start to deride the feelings, thoughts and desires of our fellow humans. Science and Scientism Finally, a word about science and scientism in psychotherapy research. We all have a need for certainty and experience discomfort with not knowing, and risk anxious retreat from ignorance into pseudo knowledge (so characteristic of the early years of medicine). A scientific approach has obviously been incredibly helpful and has saved many millions of lives. To argue against it is not just churlish; it is clearly unethical and destructive. But to argue for a mechanical reading of evidence, as some clinical psychologists have done (Chambless & Hollon, 1998; Chambless et al., 1996) equally skirts the risk of doing harm. Research evidence collected as part of the present initiative will need to be carefully weighed. Multiple channels of evaluation are needed and they need to be kept open and actively maintained. No self-respecting clinicians will change their practice overnight. They would be unwise to do so. Evidence has to be read and evaluated, placed into the context of what is possible, desirable and fits with existing opportunities. It should be remembered that in mental health, at least but also probably in most areas of clinical treatment method, accounts can be given for a relatively small proportion of the variance in outcome relative to the nature of the patients problem (Weisz, Weiss, Han, Granger, & Morton, 1995; Weisz, Weiss, Morton, Granger, & Han, 1992), which may well interact with the skills of the attending clinician. This latter form of variance is to be cherished, not only because that is where the art of medicine lies, but also because it is in the study of that variability that future major advances in health care may be made, as long as we can submit these to empirical scrutiny.

MAJOR DEPRESSION
About 20 psychodynamic psychotherapy trials have been published dealing with the treatment of depressive and anxiety disorders or symptoms (Crits-Christoph, 1992; Leichsenring, 2001). Along with other therapies, it has been shown to have better effectiveness in open trials or compared to waiting list (Shefler, Dasberg, & Ben-Shakhar, 1995) or outpatient treatment in general (Guthrie, Moorey, Margison, et al., 1999). In the light of relatively readily available alternative treatments, the critical demonstrations concern that of an equivalence, or perhaps even superiority, to alternative treatment approaches. Reviews There have been two recent relevant reviews of the literature (Churchill et al., 2001; Leichsenring, 2001). In addition the National (England and Wales) Institute for Clinical Excellence (NICE) is conducting a systematic review in order to produce guidelines for treatment of depression within the National Health Service. The Churchill review concerned treatments for depression of 20 sessions or less published up to 1998. Of the studies suitable to include in metaanalysis six involved psychodynamic therapy. Improvement was found to be over twice as likely with cognitive behaviour therapy (CBT) as it was with psychodynamic therapy. However, to conclude from this that CBT is superior to psychoanalytic psychotherapy in the treatment of depression may be premature in the light of the following considerations. (1). There was no superiority of CBT over other therapies where follow-up was available. (2). Differences between CBT and other therapies were limited in severely depressed groups. (3). A number of therapies identified in the review as psychodynamic were not bona fide therapies (Wampold, 1997). This

141

is a very important point. A In an earlier meta-analysis by Gloaguen (1998) which similarly concluded that CBT was superior to other therapies. However, the superiority of CBT could no longer be demonstrated when once interventions which were not theoretically based and had no without scientific base were removed from these comparisons, the superiority of CBT over other therapies could no longer be demonstrated (Wampold, Minami, Baskin, & Callen-Tierney, 2002). Meta-analytic reviews should not confound estimates of the effectiveness of STPP with those of non-bona fide therapies. A more positive picture apparently emerges from the review by Leichsenring (2001). This review identified six RCTs of at least 13 sessions in length studies that contrasted manualised STPP and CBT (Barkham, Rees, Shapiro et al., 1996; Elkin, 1994; Gallagher-Thompson & Steffen, 1994; Hersen, Bellack, Himmelhoch, & Thase, 1984; Shapiro et al., 1994; Shapiro et al., 1995; Shea et al., 1992; Thompson, Gallagher, & Breckenridge, 1987). The review concludes that the two forms of therapy are not substantially different as only one of the studies reviewed suggests a possible superiority of CBT. We calculate the overall Risk Ratio (RR) for the comparison to be .91 (95% Confidence Interval [CI]: .77 to 1.06) indicating that CBT is only 9% more likely to generate remission than STPP. Although not reported by Leichsenring, metaanalytic comparison of follow-up data available for these studies actually reveals a slight superiority for CBT (RR = .82, 95% CI: .70, .96). This indicates that CBT increases the chance of continued remission by 20% more than STPP . We should consider the possibility of selection bias in this review. Leichsenring includes the NIMH Collaborative Depression Trial in the meta-analysis, which is, to say the least controversial, as IPT was included as an STPP merely because the therapist was psychodynamically trained (Elkin, 1994; Shea et al., 1992). As neither the developers nor other reviewers consider IPT to be a psychodynamic therapy, it seems wiser not to include it in reviews of STPP. However, eEven if this study is excluded, the superiority of CBT over STPP remains (RR=.82, 95% CI: .70, .96). However the remaining four studies include a trial of social skills training relative to STPP (Hersen et al., 1984) and a study of CBT offered to carers (GallagherThompson & Steffen, 1994) neither of which seem relevant to the assessment of the relative effects of CBT for depression. Of the two studies remaining, one was group rather than individual therapy carried out with an older adult population (Thompson et al., 1987). The most appropriate conclusion at this stage might be that a meta-analysis of this literature is premature. Consideration of Individual Studies Contrasting Psychodynamic Therapy With Other Psychotherapies This conclusion reflects the regrettable fact that there are fewer controlled studies of psychodynamic therapy for depression than one might expect given the wide use of this treatment for this problem. In fact most investigators employ STPP as a contrast to an alternative therapy to which they have an allegiance. Not surprisingly, dynamic therapy is usually found to be significantly less effective (Covi & Lipman, 1987; Kornblith, Rehm, O'Hara, & Lamparski, 1983; McLean & Hakstian, 1979; Steuer et al., 1984). Given that tThe researchers lack of investment in the control treatment reflects a lack of equipoise, means that the findings from these investigations should be treated with great caution. Consistent with this formulation, b Better designed and implemented studies report no difference between CBT and STPP (Bellack, Hersen, & Himmelhoch, 1981; Thompson et al., 1987). Nevertheless, small sample size and design weaknesses argue against basing conclusions concerning psychodynamic therapy for depression on these investigations. Two major studies stand out as sound comparisons of a cognitive behavioural

142

and a psychodynamic approach to the treatment of depression because of their size, the quality of randomisation, the care of implementation, baseline assessment and outcome measurement, and clarity of the therapeutic interventions assessed. These are: 1) The Sheffield Psychotherapy Project and 2) The Helsinki Psychotherapy Study. The Sheffield Psychotherapy Project The Sheffield Psychotherapy Project (Barkham, Rees, Stiles et al., 1996; Shapiro et al., 1994; Shapiro et al., 1995) randomised 169 patients meeting criteria for Manic-Depressive Disorder (MDD) to CBT or STPP. 117 completed the study; 103 were followed up at one year. Treatment was brief (eight or sixteen sessions). Most would agree that eight sessions represents a sub-clinical dose of STPP. In line with this, those who received only 8 sessions of psychodynamic therapy were doing less well at one year than those with 8 sessions of CBT. At 16 sessions the two treatments appeared to be equally effective (RR=.93, 95% CI: .68, 1.27). A further important finding was that those with more severe depression at the start of treatment were less likely to maintain gains in both arms of the trial. Patients with severe depression (Beck Depression Index [BDI] score > 27) appeared to require longer treatment regardless of treatment type. About 30% of patients remained asymptomatic from post-treatment to one year follow-up. This interaction, however, was not found in a small scale community based attempt at replication (Barkham, Rees, Shapiro et al., 1996). Overall neither of these short-term treatments (CBT or STPP) appeared to be strikingly effective in the medium term but STPP performed broadly comparably to CBT. The Helsinki Psychotherapy Study The fullest assessment of psychodynamic psychotherapy for mood disorder thus far comes from the initial report of the Helsinki Psychotherapy Study (Knekt & Lindfors, 2004). This in many ways exemplary randomised study compared a problem solving-solution focused therapy (SFT) to short and long-term psychodynamic psychotherapy. The study also had an arm for full psychoanalysis, to which subjects were self-selected and additionally screened for suitability. So far only findings related to the two short-term arms of the trial are available. SFT is a relatively novel approach (Lambert, Okiishi, Finch, & Johnson, 1998) that emphasises the collaborative efforts of patient and therapist to identify a problem and find solutions to it. It is conducted once every second or third week up to a maximum of 12 sessions (the actual number of sessions in the trial was 10 over about 7.5 months). STPP, following Malans approach (Malan, 1976), was scheduled over 20 once-weekly sessions over 5-6 months (actual number of sessions in the trial was 15 over a period of about 6 months). Data was collected on auxiliary treatments such as psychotropic medication and the additional use of psychotherapy services. An unusually wide range of assessments are administered at various time points (5 assessment points in the first year) up to 60 months after baseline measurement, in order to pick up the long term and meta-symptomatic changes believed to be associated with STPP. 638 patients were referred from clinical services (not recruited for the study by advertisement or the like) of whom 381 were eligible and willing to participate in the study. 97 patients were allocated to SFT, 101 to STPP and 128 to LTPP. Over 82% of the sample met DSM criteria for depressive disorder and 43% for some type of anxiety disorder. 57% had mood disorder alone and 14% had anxiety disorder alone. The sample appears moderately severe (average GAF score was 55, and SCL-90 GSI 1.29) slightly less severe than the Sheffield sample.

143

The average Hamilton Depression (HAM-D) score was 15.7. Surprisingly, only about a quarter were on any kind of psychotropic medication. Given the careful design and implementation of the complex methodology and the exceptionally large sample size (adequate power to detect a 20% difference), the observation of an absence of significant differences between the two groups is impressive. The reduction in depression and anxiety scores was relatively large (e.g. the BDI decreased by almost 50% (from 18 to 10) and the Hamilton from 15 to 11 at 12 months. The average Global Assessment of Functioning (GAF) increased to about 65. On most of the measures the gains were apparent by 7 months and remained stable thereafter. About 30% of the 82% of patients who were suffering from a mood disorder lost the diagnosis by 7 months and changes after that were no longer statistically significant. Both forms of therapy showed about 20% of recovery from personality disorders by 7 months. This increased to 46% in the STPP group by 12 months but in SFT the percentage recovered from Personality Disorder (PD) diagnosis did not increase (RR=2.1, 95% CI: 1.36, 3.25). However, we cannot make too much of this difference given the absence of adjustments for Type I error and the great number of comparisons carried out. Measuring recovery in terms of achieving below clinical cut-point scores on the BDI, and Hamilton ratings, SFT appeared to achieve somewhat more change faster although there were again no significant differences after the 9 months time point (RRBDI@3months=.49, 95% CI: .27, 0.88). Work ability index improved for both groups but only by about 10% and this was achieved by 7 months with little change thereafter. Interestingly, days of sick leave was reduced significantly for both groups. Similarly, a range of measures on social functioning improved over the first 7 months but produced no notable change later and no differences were seen between the treatment groups. In summary, the trial demonstrated rapid and generally similar decreases on self-reported and observer rated depressive symptoms during the first months of therapy and less prominent reductions in symptoms later. Remission was somewhat more rapid with SFT and continued recovery from a PD diagnosis was more marked for STPP. The rate of recovery reported here is comparable to those found in other studies involving CBT. Broadly, the study demonstrates that a fairly generic form of brief psychoanalytic psychotherapy is as effective in the treatment of depression as a less generic but previously established as empirically supported form of CBT (Mynors-Wallis, Davies, Gray, Barbour, & Gath, 1997; Mynors-Wallis, Gath, Day, & Baker, 2000; Mynors-Wallis, Gath, Lloyd-Thomas, & Tomlinson, 1995). The study was unusual in the breadth of outcomes covered. It permitted the observation that recovery of work ability, social functioning and personality functioning appears to be far smaller than reductions of acute psychiatric symptoms (e.g. Pre-Post Standardised Mean DifferenceBDI (SMD)=9.65, 95% CI: 8.66, 10.64; Pre-Post SMDSAS(work)=3.33, 95% CI: 3.79, 12.84). Measurement in social adjustment domains does not appear to advantage the less symptom focused STPP. This is important in the interpretation of other trials. The limited gain observed in social adjustment is consistent with current models of change in psychotherapy (Howard, Lueger, Maling, & Martinovitch, 1993). The absence of a no treatment control group in this study makes it hard to estimate the proportion of observed reductions in symptoms that was not due to psychotherapy. However, since effective treatments for depression exist randomisation of patients to a purely placebo arm for a 12 months period would be unethical: adequately designed studies will inevitably lack a no-treatment control arm. The treatments were well described but not fully manualised. Adherence is not reported.

144

Consideration of Individual Studies Contrasting Psychodynamic Therapy with Pharmacotherapy When offered alone, psychotherapy and medication are of equivalent efficacy and psychotherapy is rarely superior (Roth & Fonagy, in press). There is evidence of added value when psychotherapy is added to medication but most of this evidence is from IPT and CBT studies. Two recent studies assessed adding STPP to an antidepressant regimen. A Dutch study (de Jonghe, Kool, van Aalst, Dekker, & Peen, 2001) assigned 84 patients to antidepressant and 83 to combined therapy. The pharmacotherapy protocol started with fluoxetine and allowed a switch to amitryptyline if participants were non-responsive and finally to moclobemide. The psychotherapy was a psychoanalytically oriented brief supportive therapy associated with the work of the first author (de Jonghe, Rijnierse, & Janssen, 1994). An unexpectedly large number of patients refused pharmacotherapy alone, so of the 167 patients randomised, 57 started in pharmacotherapy and 72 in combined therapy. Intent to treat remission rate in the combined psychodynamic psychotherapy and pharmacotherapy on the Hamilton Depression Rating Scale (HDRS) was 37% while it was only 15% in the pharmacotherapy sample (RRHDRS<7=2.36, 95% CI: 1.33, 4.18). The findings indicate that not only is combined STPP and pharmacotherapy accepted better but symptomatic indicators of success rate, both clinician and self rated, suggest high levels of success. The success rate of 37%, defined by HDRS less than 7, for the combined group is not impressive. Across a range of measures, however, by 24 months the success rate of the combined therapy is nearly 60% and that of pharmacotherapy alone rises to 40% (RRHDRS<7=1.5, 95% CI: 1.09, 2.06). The superiority of the combined treatment is maintained but does not appear to increase. It is not clear how combined treatment would compare with a psychotherapy alone condition. In some analyses, this has been a difficult difference to demonstrate (Thase et al., 1997). A second comparison study (Burnand, Andreoli, Kolatte, Venturini, & Rosset, 2002) randomly assigned 95 patients to a combination (clomipramine and STTP) or a clomipramine alone condition. Despite an attrition rate of 22%, the intent to treat analysis revealed somewhat fewer treatment failures for the combined treatment as well as better work adjustment scores. No group differences were noted on the HDRS however. There were more frequent instances of hospitalisation and more days spent in hospital for the clomipramine alone treatment. Combined therapy was also associated with fewer days lost from work. This, together with saving on hospitalisation costs, led to an estimated cost saving per episode of $2,300. It should be noted that the benefits of STPP were detectable even though the pharmacotherapy group had some nonspecific psychological input (attention placebo) and the psychotherapy was delivered by nurses rather than certified psychotherapists. This latter finding echoes the success of other psychodynamic interventions where the therapist is supervised by but is not a fully trained psychoanalytic psychotherapist(Bateman & Fonagy, 2001). Process Factors and Moderating Variables Support for a psychodynamic approach to the treatment of depression may also accrue from the demonstration that the inclusion of interventions specific to a psychoanalytic approach is associated with good outcome even in therapies for depression born of a different orientation. For example, in a study of CBT, the extent of focus on parental issues turned out to be positively associated with outcome (Hayes, Castonguay, & Goldfried, 1996). In another study, of CBT and STPP, the differences between the success of either intervention appeared to be correlated with the use of interventions prototypically considered psychoanalytic (Ablon & Jones, 1998). However, similar evidence can be readily mustered to support CBT. Using the same instrument (Enrico

145

Joness Psychotherapy Q-sort) on tapes of the NIMH treatment of depression trial, Ablon and Jones demonstrated the superiority of therapies where process codings of both IPT or CBT more closely resembled the CBT prototype(Ablon & Jones, 1999; 2002). Studies of psychodynamic psychotherapy process and outcome have not thus far demonstrated powerful associations between change and putative mechanism of action. For example, a negative association has been reported between the number of transference interpretations and therapy outcome, indicating that the overuse of this technique frequently regarded by clinicians as essential to therapeutic success, may even be iatrogenic(Connolly et al., 1999). Summary The current evidence base of psychodynamic therapy for depression is weak relative to the number of psychoanalytic therapists and the rate at which evidence is accumulating for other approaches. The psychodynamic approach may be marginalised, not by its relative lack of effectiveness, but by the sparseness of compelling demonstrations of its comparability to empirically supported alternatives. There is some evidence relating to brief psychodynamic therapy (up to 24 sessions) but no evidence for long-term therapy or psychoanalysis, despite the fact that data from trials of depression indicate the need for more intensive treatment. As Westen and Morrison (2001) observe, the evidence for the effectiveness of CBT for depression in the medium term is not sound. However, none of the therapies appear to differ markedly from each other. In the two cases where brief psychodynamic therapy was compared with CBT or problem solving therapy, the observed size of the effects were similar in the groups contrasted and in turn similar to results reported in other studies of CBT, IPT, and couples therapy. Broadly speaking in 4-6 months of therapy about half of those treated are likely to remit. About half of these will experience a remission in the succeeding 12 months. As Roth and Fonagy (Roth & Fonagy, in press) observe, the data are consistent with the assumption that a proportion of patients in any research sample will respond to therapeutic intervention of any kind. An appropriate future strategy for psychodynamic psychotherapy research on depression might be to compare the effectiveness of relatively long-term psychodynamic psychotherapy with alternative forms of intervention in patients who are non-responders in trials of CBT, IPT or pharmacotherapy. A further weakness of the evidence base for psychodynamic treatment studies of depression is that no STPP was tested twice by independent research groups (Chambless & Hollon, 1998) and most manuals subjected to systematic inquiry are idiosyncratic, and their testing is restricted to the location in which they were developed.

ANXIETY DISORDERS
Research on anxiety disorders is normally subdivided into research on phobia, generalised anxiety disorder, panic disorder (with and without agoraphobia), post-traumatic stress disorder (PTSD) and obsessive-compulsive disorder (OCD). These, often co-morbid with depression (Brown, Campbell, Lehman, Grisham, & Mancill, 2001), are the most commonly encountered disorders either in community surveys or in primary mental health services. Anxiety disorders are central to psychoanalytic theory (Milrod, Cooper, & Shear, 2001) and are probably the most common presenting complaints in psychodynamic therapeutic practice. Disappointingly, for at least two of the most common anxiety problems (social phobias and specific phobias) there are no diagnosis-specific controlled trials of psychodynamic therapy. The field is dominated by CBT packages that combine a range of approaches with almost no studies of non-behavioural

146

approaches except for a small trial of interpersonal therapy (Lipsitz, Markowitz, Cherry, & Fyer, 1999). Generalised Anxiety Disorder Treatments developed for generalised anxiety disorder are dominated by anxiety management (relaxation, positive self talk) and cognitive therapy focusing on identifying and modifying worrying thoughts. Because of the sparsity of the literature, meta-analytic and systematic reviews are not very informative. Westen and Morrisons meta-analysis (Westen & Morrison, 2001) identify a study of psychodynamic therapy (supportive expressive therapy) (CritsChristoph, Connolly, Azarian, Crits-Christoph, & Shappell, 1996) but does not address differences in efficacy across the four types of therapy examined. Another meta-analysis (Fisher & Durham, 1999) establishes a clinical cut-off on the Spielberger Trait Anxiety Scale, to identify the proportion of participants who recover at post-therapy, remain in remission at six months, as well as those who recover in the six months post-therapy period. By these criteria, overall, at six months 36% are unchanged, 24% are improved and 38% are recovered. Only 4% of patients are considered as recovered following psychodynamic therapy. But this is based on just one study (Durham et al., 1994). This contrasts with 60% for individual applied relaxation therapy and 51% for CBT. Notably, non-directive therapy was shown to have recovery rates of 38%, suggesting that neither relaxation nor restructuring of worrying thoughts are necessary for substantial improvement. However, the review includes too few trials to be a definitive statement about the relative effectiveness of these therapies in anxiety. A number of individual studies of psychodynamic psychotherapy have provided some relevant data. Two open trials have been reported. Crits-Christoph and colleagues (Crits-Christoph et al., 1996) followed 26 patients for a year. The patients had 16 once-weekly sessions followed by booster sessions once every three months. At post-therapy 79% no longer met diagnostic criteria for Generalized Anxiety Disorder (GAD). A second study by Durham and colleagues (Durham et al., 1999) contrasted analytically based psychotherapy with CBT and anxiety management training in an RCT. 99 patients with a GAD diagnosis for at least 6 months (mean duration 30 months) were randomly assigned to either high frequency or low frequency (weekly or fortnightly) CBT or analytic psychotherapy. Unfortunately, the groups ended up poorly matched with the CBT patients in the low frequency condition significantly less severely affected. Importantly, the gains from psychoanalytic psychotherapy were less than those for cognitive therapy and were also less likely to be maintained at one-year follow-up. Thus, whilst 60% of patients treated by CBT met clinically significant change criteria at one-year follow-up, only 14% of those in analytic psychotherapy did so (RRHARS=.11, 95% CI: .03, .37). At this stage higher frequency was associated with better maintenance of gains for both CBT and analytic psychotherapy. However, when the sample was followed-up almost a decade after the end of treatment (Durham, Chambers, MacDonald, Power, & Major, 2003) only about half of those who had achieved recovery at 6 months maintained their gains in the very long term. By this stage there was no difference in outcome between CBT and non-CBT treated participants. A major limitation of this trial, but also a helpful pointer, was the nature of the psychoanalytic therapy offered. The two therapists, although both experienced psychiatrists, were training in psychoanalysis and are not reported to have had special training in brief therapeutic methods, and had no manual to follow. Under these circumstances it is highly likely that the therapy delivered was not an effective version of STPP. If the effectiveness of cognitive therapy

147

were assessed on the basis of two individuals trained in generic CBT, without further training specific to the trial, this would generally be considered unacceptable. It is possible, as the techniques used are not specified in the report, that these therapists used inappropriate techniques borrowed from the long-term therapy in which they were currently training and unhelpfully applied these in a time limited context. Panic Disorder As with GAD, non-directive psychodynamic therapies have been used as control treatments in a number of CBT trials (e.g.Beck, Sokol, Clark, Berchick, & Wright, 1992; Craske, Maidenberg, & Bystritsky, 1995). The relative absence of therapeutic equipoise makes these trials suspect and evidence from them can have few implications for the effectiveness of short-term psychodynamic treatments. When more effort is made to create a credible placebo, differences between treatment and control groups are reduced or disappear completely. For example, one study randomised 45 patients with panic disorder to 15 sessions of CBT or a therapy described as non-prescriptive where the therapist was encouraged to offer reflective listening (Shear, Pilkonis, Cloitre, & Leon, 1994). No significant differences were found at either post-therapy or at six months follow-up. In a partial replication, the same group (Shear, Houck, Greeno, & Masters, 2001) did find a difference between CBT and this non-prescriptive form of therapy, but the difference is small relative to less credible control interventions. In the meta-analysis by Nordhus and Pallesen (2003) of psychological treatments for later life anxiety, a number of studies are noted where the effectiveness of CBT is reduced or even reversed when the control condition is a highly credible psychotherapeutic placebo (e.g. Wetherell, Gatz, & Craske, 2003). There are few trials of brief psychodynamic therapy for panic and none unequivocally addresses the effectiveness of this approach. Milrod and her colleagues have worked over a number of years to establish the evidence base for a manualised panic focused psychodynamic psychotherapy (Busch, Milrod, Cooper, & Shapiro, 1996; Busch, Milrod, & Singer, 2000; Milrod, Busch, Cooper, & Shapiro, 1997). This team report an open trial of psychoanalytic psychotherapy with 21 patients seen twice weekly over 12 weeks and with a six month follow-up (Milrod, Busch et al., 2001; Milrod et al., 2000). 16 of the 21 participants showed remission, (defined as a reduction from baseline of more than 50% on the panic disorder severity scale). The results proved stable at an unusually long follow-up period of 40 weeks. Of those completing treatment, 93% were considered remitted at the end of treatment and 90% at follow-up. The study is in many ways exemplary. In particular, there was careful attention to measurement issues, careful training of therapists and attention to treatment integrity. It is also remarkable for the involvement of senior psychoanalysts in the design and delivery of this structured and symptom-focused but unequivocally psychoanalytic treatment. Although the effect sizes are comparable to those observed in the best trials of cognitive behavioural therapy (e.g.Barlow, Gorman, Shear, & Woods, 2000), the absence of a control group and relatively small sample size limits the generalisability of the conclusions. Wiborg and Dahl (Wiborg & Dahl, 1996) reported a controlled study that examined the effect of adding psychodynamic therapy to treatment with clomipramine. Thirty patients were randomised to clomipramine or clomipramine with 15 weeks of manualised dynamic psychotherapy based on Davenloo, Malan, and Strupp and Binders work. There was a follow-up at 6, 12 and 18 months of treatment. At the end of treatment all patients in the combined treatment group were free of panic attacks (RRHRSD=1.33, 95% CI: 1.04, 1.72), but at 6 months follow up all clomipramine-treated subjects were panic free. By 18 months 75% of the drug only group had relapsed but only 20% of

148

the group who had received psychotherapy (RRHRSD=3.2, 95% CI: 1.45, 7.05). The differences were apparent even when adjustments were made for initial severity of symptoms and social adjustment. Further advantages were observed in terms of the psychotherapy group reporting fewer side effects from the medication. This study provides evidence that STPP is an effective adjunct to pharmacotherapy in panic disorder, but of course cannot speak to the effectiveness of this treatment in the absence of clomipramine. It is also important to note that clomipramine was administered by the treating psychotherapist in the experimental arm of the trial but by a nonpsychiatrically trained general physician in the control group. The findings confirm the results of an early trial (Klein, Zitrin, Woerner, & Ross, 1983) where the effect of imipramine for phobic problems was (to the authors surprise) as strongly enhanced by STPP as by behaviour therapy (RR=.97, 95% CI: .66, 1.43). As described above, the Helsinki Psychotherapy Study (Knekt & Lindfors, 2004) included a significant number of individuals diagnosed with panic disorder (n=34) and generalised anxiety disorder (n=37). In addition, the authors observed that practically all the patients (more than 95%) had Hamilton Anxiety scores higher than 7 at baseline. There was a slight, but not statistically significant, difference between the groups in terms of the impact of the treatment on anxiety problems: 56% of those with anxiety disorders lost their diagnosis by seven months of STTP, compared with 42% of those in SFT (RRHARS=1.34, 95% CI: 1.0, 1.78). By 12 months the difference had narrowed to 62% and 52% respectively (RRHARS=1.19, 95% CI: .93, 1.51). Whilst this difference is not significant, it underscores the potential for STPP to rapidly assist in the problems of anxiety as well as depression. There was a 34% decrease in the Hamilton anxiety rating scale for STPP and a 28% decrease for SFT at 12 months, but almost all of this had been achieved by seven months. Stress Related Conditions: Post Traumatic Stress Disorder (PTSD) and Complex Grief Reaction Psychodynamic approaches to PTSD focus on the meaning of the traumatic event for the persons sense of self and their place in the outside world (Horowitz, Marmar, Weiss, DeWitt, & Rosenbaum, 1984). Studies supporting this approach are either case reports or open trials. Open trials of the psychodynamic treatment of female victims of sexual assault have tended to be positive but methodologically problematic (Cryer & Beutler, 1980; Perl, Westlin, & Peterson, 1985; Roth, Dye, & Lebowitz, 1988). Scarvalone and colleagues (Scarvalone, Cloitre, & Difede, 1995) contrasted a psychodynamic group therapy with a wait list control group. All participants (N=40) had histories of sexual abuse although not all had current symptoms of PTSD. After treatment 39% of those in the psychodynamic group as opposed to 83% of those in the control group met diagnostic criteria. (RRDPTSD=.47, 95% CI: .27, .83). A relatively large randomised controlled trial (RCT) from Holland (Brom, Kleber, & Defares, 1989) contrasted psychodynamic therapy, hypnotherapy and trauma desensitisation in the treatment of 114 individuals with PTSD diagnosis. Many were bereaved and only about 20% reported experiencing a traumatic event. There was also a wait list control group. Mean length of treatment varied somewhat between conditions (trauma desensitization=15 weeks, hypnotherapy=14.4 weeks, STPP=18.8 weeks). The study reports all treatments to be superior to the waiting list control group with clinically significant improvements in about 60% of treated patients and 26% of the untreated (RRDPTSD=2.28, 95% CI: 1.12, 4.64). At post-treatment psychodynamic therapy seemed to have the weakest effects, but therapeutic changes continued in

149

this group and by follow-up they matched or exceeded those of other therapies. Interestingly, trauma desensitisation had stronger influence on intrusions and psychodynamic therapy had more influence on avoidance. Complex grief reaction was the focus of a further RCT of brief individual psychotherapy contrasted with a group therapy run by non-clinician volunteers who were experts by experience (Marmar, Horowitz, Weiss, Wilner, & Kaltreider, 1988). The therapy was based on Horowitzs model of pathological bereavement and was delivered by experienced therapists. 61 patients were randomised and outcomes were assessed at the end of treatment, at four months and one year follow-ups. Patients received 12 sessions of therapy in each group. The superiority of the psychotherapy group was principally in terms of reducing attrition in the early and late phases of the treatment. Nearly a third of those in group treatment terminated in the first third of the treatment (RRDROPOUT=.42, 95% CI: .24, .73). Self report results provided evidence of the superiority of psychotherapy at four months and one year on the general severity index of the SCL90 (SMDSCL90=-.6, 95% CI: -1.1, -0.8). Other than the GSI, observer ratings and self-report differences favoured the psychotherapy group but were not statistically significant. This was most likely because the self-help group was surprisingly effective, at least for those who attended them. This trial was designed before the wider recognition of the value of self-help groups, otherwise perhaps the investigators might have chosen a less challenging comparison. An impressive programme of work by William Piper and colleagues (Piper et al., 1991; Piper, Joyce, McCallum, & Azim, 1998) examined the effects of time-limited interpretive therapy, in both an individual and in a partial hospitalisation context (Piper, 1996). However, these studies were conducted with heterogeneous samples of psychiatric outpatients and therefore are not relevant to a diagnosis based review. An important trial explored the value of interpretive versus supportive group therapy for individuals with complicated grief reactions(Ogrodniczuk, Piper, McCallum, Joyce, & Rosie, 2002; Piper, McCallum, Joyce, Rosie, & Ogrodniczuk, 2001). Both therapies were modified to be appropriate for group treatment of grief, but while interpretive therapy was aimed to enhance insight about repetitive conflicts associated with the losses, supportive therapy included praise and gratification. The treatments were manualised, sessions were videotaped and the groups met 12 times for 90 minutes once a week. Both treatments were effective in terms of an exceptionally broad range of outcome measures, covering general symptoms, grief symptoms and target problems. The most informative finding to emerge from this study was the interaction between type of therapy and a measure of object relatedness. Psychological-mindedness was associated with improvement in both interpretive and supportive therapy. An intriguing and complex interview-based measure of the quality of object relations, the QOR (Piper et al., 1991), was found to interact with these modes of therapy in a fashion meaningfully related to psychodynamic formulations. High quality or mature object relations on the QOR appear to predict greater benefit from interpretive therapy whilst those whose relationships are judged to be more primitive, searching, or controlling were more likely to benefit from supportive therapy. Subsequent investigation (Piper, Ogrodniczuk, McCallum, Joyce, & Rosie, 2003) suggested that the balance of positive and negative affect expressions during therapy was the important mediator of this association. This well conducted programme suggests that STPP may be a relatively effective treatment for complex grief reactions and that individuals with more mature representations of interpersonal relationships are most likely to benefit.

150

Summary Anxiety treatment research represents the home base of cognitive behavioural approaches. GAD represents its most significant challenge. One uncontrolled trial suggested that STPP might have something to offer these patients. A controlled trial comparing an unspecified form of STPP to cognitive therapy showed the latter to be substantially more effective in the short and medium term. The disappointing results from the psychodynamic arm of the trial may relate to the non-specificity and unstructured nature of this therapy. Panic attacks appear to be relatively well treated by 15 to 20 sessions of CBT. There is a promising psychodynamic therapeutic approach to panic that might match CBT in efficacy which requires extension and replication in multi-centre controlled trials. The superiority of CBT over other approaches is probably limited, as shown by reduced effect sizes in controlled trials that have active placebo treatments. The challenge for a psychodynamic approach is to identify a way to address limitations in CBT, either in terms of long-term efficacy (Milrod & Busch, 1996) or a more pervasive impact on social functioning. Interestingly, evidence from the Helsinki trial, where this was a focus of investigation, did not support the view that short-term psychodynamic therapy had a wider impact on social functioning than problem-focused treatments. Evidence on the treatment of PTSD is also sparse, notwithstanding the central involvement of psychoanalytic clinicians in mapping the consequences of childhood and later trauma. Available controlled studies concern complicated grief and bereavement reactions, and not exposure to trauma. Nonetheless, findings from such trials are generally positive although by no means uniquely effective. It is striking that little research has been done to establish the pertinence of psychodynamic approaches to anxiety, which is so central to both psychoanalytic theory and practice, . Possibly psychodynamic therapists do not consider anxiety symptoms important enough as Freuds term signal anxiety might suggest. Bypassing the surface problem of the anxiety symptoms they attempt to achieve change in the underlying structures even in brief therapies. Losing focus they find change relatively difficult to bring about. It requires an approach such as Milrods, which retains focus on the symptom at the same time as exploring unconscious determinants, to achieve rapid change. The importance of anxiety-related problems demands that further studies should be initiated.

EATING DISORDERS
Anorexia Nervosa Most current treatment approaches for anorexia nervosa (Behaviour Therapy [BT], CBT, family therapy, psychodynamic psychotherapy) recognise the importance of establishing an appropriate dietary regime which is common to all treatments. An early study (Hall & Crisp, 1987), assigned 30 anorexic patients to either dietary advice or psychotherapy. The psychotherapy was psychodynamic, delivered either individually or involving the whole family for 12 sessions at two weekly intervals. The dietary advice group entailed 15 one-hour sessions at weekly or fortnightly intervals with a dietician. There was no difference between the groups on bodyweight at post-treatment or at one-year follow-up. Those receiving psychotherapy showed better social and sexual adjustment scores. The sample sizes made it difficult to interpret these results, as the majority of the participants in the anorexic arm showed substantial weight gain and the inferior mean relative to the control groups was due to three individuals who showed a substantial weight loss. Needless to say, in this early study the psychotherapy (STPP) was not manualised. An influential study by Russell and colleagues (Russell, Szmukler, Dare, & Eisler, 1987) contrasted family and individual therapy for 80 inpatients, the majority of whom (57) were

151

anorexic. The therapy was not strictly psychodynamic as it included cognitive and strategic techniques. Broadly, the majority of patients in both groups had poor outcome at one year (61%), with just under a quarter having good outcomes. Whilst indicating the general difficulty of treating this group of patients, this study also suggested that late-onset patients (onset after age 19) did better with individual therapy. This finding was partially confirmed in a five-year follow-up (Eisler et al., 1997). The superiority of family therapy is restricted to anorexics with early onset and relatively short history. The largest trial involving STPP is that of Dare and colleagues (Dare, Eisler, Russell, Treasure, & Dodge, 2001). 84 patients were recruited to STPP based on Malans approach all conducted by experienced therapists, family therapy (Dare & Eisler, 1997) and cognitive analytic therapy (Ryle, 1990). The control group received standard psychiatric care (from trainee psychiatrists). This trial was carried out in a tertiary care setting; most patients had had one or more treatment failures. Not surprisingly, outcomes were poor (mean weight gains were small and left patients with under-nutrition), but provided evidence for all three treatment approaches. After one year of treatment about a third of the patients in the three specialist psychotherapies no longer met DSM criteria for anorexia nervosa as compared with only 5% of those in routine treatment (RRDAN=7.0, 95% CI: 1.0, 48.9). It was not possible to differentiate clearly between the three specialised psychotherapies in terms of an improvement based on weight gain, regularity of menstruation and bulimic symptoms although the odds ratio for improvement in STPP and family therapy was better and significantly different from routine treatment (RRFOCAL=1.99, 95% CI: .85, 4.68) (RRFAMTHER=1.55, 95% CI: .63, 3.84). The study was however under-powered and follow-up was only partial. Crisp and colleagues (Crisp et al., 1991) report a RCT with 20 patients allocated to each of four treatment arms: inpatient treatment, outpatient individual and family psychotherapy (12 sessions over 10 months), 10 group therapy sessions and initial assessment along with treatment as usual in the community. The individual therapy group was significantly better than both the inpatient and the minimal treatment group at the end of treatment. The individual psychotherapy was structured, somewhat eclectic in orientation, and delivered by experienced psychotherapists. The inpatient group received a range of treatments based on behavioural techniques coupled with milieu treatment and individual and family therapy. Gowers and colleagues (Gowers, Norton, Halek, & Crisp, 1994) reported on the outcome of the outpatient group compared with the minimal treatment group two years after the end of treatment. The outpatient group maintained and further improved their weight gain which was more than twice that of the control group. While 60% of the psychotherapy group were judged to be well (within 15% of Mean Matched Population Weight [MMPW], normal menstruation and eating habits) or nearly well (within 15% of MMPW, nearly normal menstruation and/or abnormal eating habits) only 20% of the control group were judged to be so (RRDAN=3.0, 95% CI: 1.16, 7.73). Despite the small sample size, the study suggests that substantial benefits accrue to anorexic patients from psychodynamic therapy supported by family therapy. However, the form of therapy offered was eclectic rather than a pure form of STPP; therefore effective elements may actually be associated with the strategic or directive elements. Bulimia Nervosa The majority of trials in the literature (around 30) employed either behavioural or cognitive-behavioural techniques (Hay & Bacaltchuk, 2001; Thompson-Brenner et al., 2003). Intent to treat recovery rates are only around 33%, with an advantage to individual therapies and no clear superiority to CBT over alternative active psychological interventions. No non-CBT studies

152

report long term follow-up and meta-analytic studies show a tendency towards deterioration in effect size of follow-up samples relative to post therapy observations. The meta-analyses do not speak to the specific efficacy of psychodynamic therapies. Garner and colleagues (Garner, Rockert, Davis, & Garner, 1993) reported a study contrasting STPP based on Luborskys supportive expressive model (Luborsky, 1984) with CBT. 60 patients were randomised to 19 individual treatment sessions over an 18-week period. Five patients withdrew from each treatment arm. The two treatments were equally effective in their impact for binge frequency. CBT was somewhat (but not statistically significantly) more effective in reducing vomiting frequency and significantly more effective in reducing depression. Both treatments are considered by the authors to be effective, but where differences emerged these favoured CBT. No follow-up is reported for the study, which would be critical in this context. A different psychodynamic approach was taken in a study of 33 bulimic patients (Bachar, Latzer, Kreitler, & Berry, 1999) assigned to either nutritional counselling alone or in conjunction with a form of cognitive therapy or a form of self-psychological dynamic therapy. A range of measures was used including measures of symptomatology, attitudes to food, self structure and general psychiatric symptoms. The groups were inappropriately small and preclude comments on relative efficacy. However, pre-post effects appear to be substantial for psychodynamic treatment, less notable for cognitive therapy and almost negligible for nutritional guidance. The study is further weakened by the mixture of anorexic and bulimic patients identified and the non-standard cognitive approach applied. Nevertheless it suggests psychodynamic treatment may have significantly better outcomes compared to an almost untreated group. Walsh (Walsh et al., 1997) reported an impressive investigation that contrasted psychodynamically oriented supportive psychotherapy with CBT and antidepressant medication (desipramine followed by fluoxetine). 120 women with bulimia were randomised to five treatment arms. In this trial CBT was superior to psychodynamic psychotherapy in reducing bulimic symptoms. Supportive psychotherapy appeared to offer little additional benefit to medication alone but CBT and medication combined produced greater improvement than medication alone. The results suggest that supportive psychotherapy is not particularly helpful for bulimia. Obesity Beutel and colleagues (Beutel, Thiede, Wiltlink, & Sobez, 2001) reported an unusual trial involving inpatient treatment of obesity (Body Mass Index [BMI] >35kg/m2 ) . Ninety-eight consecutive patients were randomly allocated to 6 weeks inpatient behavioural (n=46) or psychodynamic (n=52) treatments entailing individual and group work in both approaches. They measured weight loss, eating behaviour, body image and life satisfaction. While both interventions were associated with substantial gains, there were no differences between the two programs in terms of weight loss or eating behaviour changes. Summary There have been four trials of psychodynamic psychotherapy for anorexia nervosa, all of which found it to be as effective as other treatments, including intensive behavioural and strategic family therapy. None of the trials were powered adequately to distinguish conclusively between alternative treatments. Taking the results together it seems that relative to TAU, psychodynamic therapy for anorexia nervosa holds its own. The trials were performed in two London specialist units, but the particular brands of psychoanalytic psychotherapy practised were not comparable, so they cannot be considered replications.

153

STPP fares less well in the treatment of bulimia. One trial indicated that STPP was somewhat less effective than CBT, while in another study the superiority of STPP is based on a tiny sample size and an unusual implementation of cognitive therapy. In a trial exploring combined pharmacological and psychosocial treatments, non-specific supportive STPP turned out to be less effective than CBT in enhancing the effect of medication. Overall, as in other contexts, when STPP is modified for a specific clinical problem it is far more likely to be effective. It is comparable to a similarly refined cognitive behavioural approach. As a generic supportive treatment it is unlikely to be an appropriate recommendation for any of the eating disorders considered, but as a specific approach it is perhaps more likely to be of benefit.

SUBSTANCE MISUSE
Psychoanalytic therapies do not have a strong tradition in the treatment of substance misuse. There have been theoretical and clinical investigations of the problems but few substantive case series studies (Hopper, 1995; Johnson, 1999; Radford, Wiseberg, & Yorke, 1972). However, in the light of emerging data concerning the prevalence rates of alcohol and drug dependence, perhaps in excess of 7% (Hickman et al., 1999; Kessler et al., 1994; Kraus et al., 2003), it seems important to establish whether psychoanalytic psychotherapy has something to contribute to this major public health problem. Alcohol Misuse A research group led by William Miller has produced an exhaustive and periodically updated report concerning the relative effectiveness of a range of psychosocial treatments for alcohol dependence (Miller & Wilbourne, 2002; Miller, Wilbourne, & Hettema, 2003). Using a simple but relatively robust (Finney & Monahan, 1996) method of identifying empirical support in terms of the number of studies in the literature yielding positive outcomes, they identified brief interventions and motivational enhancement as best supported, followed by community reinforcement, and bibliotherapy, then various behavioural interventions (contracting, self control etc), followed by social skills. Psychotherapy is placed last in a long line. Indeed, evidence for the efficacy of psychotherapy and counselling is very limited. An early study (Levinson & Sereny, 1969) described assigning inpatients to a generic insight oriented therapy, with additional educational sessions or treatment as usual, which at the time included recreational therapy and occupational therapy. At one-year follow-up no differences were observed in terms of drinking behaviour, with somewhat greater improvements being reported from the control group. Millers database contains other similar examples (e.g.Pattison, Brissenden, & Wohl, 1967; Tomsovic, 1970). Looking at more specific dynamic psychotherapies, there are some studies of dynamic therapy which show superiority to a no treatment control group (Brandsma, Maultsby, & Welsh, 1980; Kissin & Gross, 1968). However, when the contrast is with minimal intervention approaches there appears to be little demonstrable benefit from dynamic psychotherapeutic treatment (Crumbach & Carr, 1979; Zimberg, 1974). Studies which do suggest differences (e.g.Pomerleau, Pertshuk, Adkins, & d'Aquili, 1978) are characterised by poor methodology (Miller & Hester, 1986). There is only one trial (amongst the 381 trials listed in the Miller database) that supports a psychodynamic approach (Sandahl, Herlittz, Ahlin, & Rnnberg, 1998). Investigators randomised 49 patients meeting criteria for alcohol dependence to group psychodynamic or group cognitivebehavioural therapy. All patients had already completed inpatient treatment. There were no statistically significant differences between the groups but at post therapy and more importantly at

154

15 months follow-up both groups improved. Whilst not statistically significant, the tendency was for the psychodynamic psychotherapy group to show better maintenance of gains. Even with the most effective treatment, the prognosis for alcohol dependence is not good, particularly for patients with greater chronicity. There is insufficient evidence in the literature to support psychodynamic psychotherapy as an adequate first-line therapy for patients with alcohol dependency problems. Given the comorbidity of alcohol dependence with other psychiatric problems, STPP may possibly be helpful in dealing with residual psychiatric disorder after alcohol dependence has been addressed. This suggestion will need to be subjected to empirical tests. Cocaine Dependency The outcome of cocaine abuse from naturalistic studies is similar to alcohol. A study of the effectiveness of community treatment (Simpson, Joe, Fletcher, Hubbard, & Anglin, 1999) revealed that 90 day relapse rates for long-term residential treatment were 15%, for short-term in-patient treatment 38% and out-patient programmes 29%. Given the under-reporting of cocaine use, the likely figures of relapse are 10-20% higher (Simpson, Joe, & Broome, 2002). Kang and colleagues (Kang et al., 1991) reported a major trial to study the efficacy of once-weekly psychotherapy or family therapy contrasted with group therapy led by a paraprofessional among patients with cocaine use disorders. Participants were recruited from those seeking outpatient treatment and 168 consented to be randomised to one of three arms in the trial. 122 of those in the trial were interviewed 6 and 12 months later and addiction severity indexes were compared. Attrition was extremely high with only about 50% attending more than one session and 22% more than six sessions. 19% of those who attended more than 3 sessions reported abstinence at six months. There was a significant impact for the cohort as a whole. This reflected the 19% of the 122 subjects who were no longer using cocaine at follow-up. There was a strong relationship between achieving abstinence and improvements in psychiatric symptoms and family problems. There was no relationship between attendance at therapy sessions and outcome. The results of the trial were considered by the author to indicate that outpatient once weekly psychotherapy is an insufficient treatment for cocaine use disorder. The results indicate that inpatient treatment or more intensive outpatient treatment may be required. A second trial, the National Institute for Drug Abuse Collaborative Cocaine Study (CritsChristoph et al., 1997; Crits-Christoph et al., 1999; Crits-Christoph et al., 2001) randomised 487 severely cocaine dependent patients, who met criteria for cocaine misuse with three quarters smoking crack cocaine and 33% also meeting criteria for alcohol dependence, to one of four treatments: (1). group drug counselling (GDC) following the 12 steps model; (2). GDC combined with individual drug counselling (IDC); (3). GDC combined with CBT, or (4).GDC combined with STPP. Treatments were offered over six months during which participants had 24 group sessions and a maximum of 36 individual sessions. The treatment lacked acceptability. By month three 50% of the participants had left treatment and overall only 28% completed. All treatments showed significant improvements from baseline to post-baseline in cocaine use (past 30 days). The greatest improvement was with IDC. By month six, an estimated 40% of available patients in the IDC group reported cocaine use compared with 58% in CBT, 50% in STPP and 52% in GDC alone (RRIDCvsSTPP@6months=1.24, 95% CI: .94, 1.64) (RRIDCvsCBT@6months=1.35, 95% CI: 1.03, 1.77). At 12 months follow up these percentages were 40%, 46%, 48% and 47%. The difference between IDC and other treatments was statistically significant. CBT and STPP retained patients better, but IDC produced more improvement in terms of abstinence. Interestingly, higher levels of alliance were associated with better retention in IDC and STPP but with worse retention in CBT (Crits-Christoph

155

et al., 2001). There was no difference between the treatments in terms of associated psychological, social and interpersonal measures (Crits-Christoph et al., 2001). It should be noted that, whilst this is one of the most sophisticated tests of the value of STPP, there are some crucial flaws in the design. In the extended (2 weeks) assessment 46% of the initial participants (those with the highest level of dependency needs) were lost to the trial, and patients on psychotropic medication who might have benefited most from STPP or CBT were excluded. Opiate Dependency Meta-analyses of the treatment of opiate dependency (Brewer, Catalano, Haggerty, Gainey, & Fleming, 1998; Prendergast, Podus, Chang, & Urada, 2002) identified short treatment length, low treatment integrity, and low levels of staff training to be associated with relatively poor outcome. Meta-analyses reveal effect sizes to be relatively low (.3) and, unusually, fail to identify a particular treatment type as especially beneficial. Woody and colleagues (Woody, McLellan, Luborsky, & O'Brien, 1987; Woody et al., 1983; Woody, McLellan, Luborsky, & O'Brien, 1990; Woody, McLellan, Luborsky, & O'Brien, 1995) performed a crucial randomised study of supportive expressive therapy contrasted with CBT in opiate dependent individuals on methadone. These two forms of therapy were compared with 24 weeks of drug counselling alone. The numbers of sessions for the three conditions were not equal: 17 for drug counselling alone, 12 for STPP and 10 for CBT. Of 305 patients who met criteria, 185 of these agreed to take part but only 110 engaged in the trial. Patients in all three groups showed improvements in lessened drug use, criminal behaviour, and psychological function at 7 months follow-up. There were advantages to the STPP group in psychological problems, days worked, money legally earned, drug use and advantages to CBT in dealing with legal problems. Both therapy groups did better than drug counselling only group. At 12 months follow-up data were available from 93 participants with the two psychotherapy groups showing more improvement. CBT showed the same kind of advantages on most measures as STPP. However 44% of STPP vs 26% in CBT and 18% in drug counselling alone were off methadone at 12 months (Woody et al., 1987) (RRSTPPVSCBTMETHADONE=1.71, 95% CI: .88, 3.31) (RRSTPPVSCONTROLMETHADONE=2.44, 95% CI: 1.12, 5.3). A partial replication of this study (Woody et al., 1995) in a community setting addressed a limitation of the previous study that psychotherapy participants had both a drug counsellor and a psychotherapist. In the replication study the clients randomised to drug counselling alone (n=41) had access to two drug counsellors, matching the psychotherapy group (n=82). The number of sessions between the arms were also equated in the community replication. There were no differences between the groups in positive opiate urine samples but there was a difference as far as cocaine urine analysis was concerned (22% in STPP and 36% in DC (RRCOCAINESTPPVSDC=.6, 95% CI: .34, 1.06). The pattern of results across time consistently indicated that while the groups were equivalent at the end of treatment the drug counselling alone group worsened during follow-up whilst the STPP group improved over the same period. study however encountered the problems that face effectiveness as opposed to efficacy studies (rapid staff changes, sub-optimal clinical protocols, lack of co-operation from clinical sites). Thus the degree of change observed is even more impressive. Summary Many studies considered the efficacy of brief interventions for alcohol dependence. For alcohol problems of low severity brief interventions seem to be the interventions of choice.

156

Psychodynamic psychotherapy along with other formal psychological therapies appears not to be particularly helpful when offered as a stand-alone treatment. On the whole, successful interventions appear to be targeted at drinking behaviour and testable psychodynamic protocols of this kind have not yet been developed. Again, for low levels of cocaine dependency, briefer treatments appear to be appropriate. But for individuals with more severe problems, both engaging with treatment and maintaining commitment to formal psychotherapy appears problematic. Supportive expressive psychotherapy appears of almost no value in the context of cocaine misuse. In fact treatments that do not engage with clients in the community context appear to be of limited relevance. It is an obvious question as to whether STPP could be modified to incorporate community involvement. A different picture emerges in the context of opiate abuse where psychodynamic treatment was shown to be efficacious in two trials, unfortunately (from the standpoint of EST criteria) carried out by the same team. However, in this context, there is a prima facie case for the unique effectiveness of supportive expressive therapy as neither IPT (Rounsaville, Glazer, Wilber, Weissman, & Kleber, 1983) nor certain cognitive therapies (Dawe et al., 1993; Kasvikis, Bradley, Powell, Marks, & Gray, 1991) appear to have quite the same impact. Nevertheless, generic counselling or certain types of family-based interventions may enhance the effectiveness of methadone treatment just as STPP appears to. In this area there is urgent need for replication by an independent group of workers willing to implement the supportive-expressive therapeutic strategy. If a place is to be found for psychodynamic psychotherapy in substance abuse protocols it is unlikely to be in offering formal therapy as a primary treatment. Rather, taking the lead from the opiate work, a niche needs to be found where psychodynamic intervention provides appropriate support for what is ultimately a physical dependency requiring physical treatment rather than hoping that a psychological intervention by itself is capable of resolving a physical dependency. There is urgent need to identify protocols that sequence traditional forms of psychosocial treatments with interventions for physical dependence within a single integrated package.

PERSONALITY DISORDERS
Personality disorders represent a special challenge for outcomes research because of the high level of comorbidity between Axis I and Axis II diagnoses and within Axis II diagnoses (Swartz, Blazer, & Winfield, 1990; Zimmerman & Coryell, 1990). Treatment research is somewhat limited, powerfully enhanced by recent activity in new approaches to cognitive behavioural therapy (Blum, Pfohl, John, Monahan, & Black, 2002; Koons et al., 2001; Linehan et al., 1999) as well as psychodynamic (Bateman & Fonagy, 2004; Clarkin et al., 2001; Clarkin, Levy, Lenzenweger, & Kernberg, 2004b; Ryle & Golynkina, 2000) treatments. There have been two meta-analyses of psychological therapies. Perry and colleagues (Perry, Banon, & Ianni, 1999) identified 15 studies, including six randomised trials. Substantial effect sizes were identified pre- to post- treatment (E.S. = 1.1-1.3) which reduced to around .7 in studies where active control treatments were used. A more focused meta-analysis (Leichsenring & Leibing, 2003) considered only trials which used either CBT or psychodynamic therapy and identified 22 studies, 11 of which were RCTs. Pre-post effect size for psychodynamic therapy was 1.31 based on 8 studies and for CBT was .95 based on 4 studies. There was an insignificant correlation between treatment length and outcome. The limited number of studies, compounded by heterogeneity of clinical populations and methods applied, suggests that meta-analysis at this stage may be premature. Further, many of the studies included in these meta-analyses did not have the aim of treating Axis II disorders.

157

Notwithstanding these limitations, the broad conclusion from these aggregated figures would be that CBT and psychodynamic therapy are equally effective. Borderline Personality Disorder (BPD) There are more studies of borderline personality disorder than of other PDs. There have been a number of uncontrolled open trials of the psychodynamic treatment of borderline personality disorder. The Menninger Study of 42 patients carried out in the 1950s is historically important as the first serious, methodologically relatively sound, attempt to evaluate the outcome of any type of psychological therapy (Wallerstein, 1986, 1989). It was a study of psychoanalysis and expressive or supportive psychodynamic psychotherapy. One may well ask what happened to this pioneering spirit? The studys findings are complex but broadly imply that more mature personalities with better interpersonal relationships responded well to expressive interpretive therapy, whereas those with low ego strength responded better to supportive interventions. There have been a number of other naturalistic studies (Antikainen, Hintikka, Lehtonen, Koponen, & Arstila, 1995; Karterud et al., 1992; Monsen, Odland, Faugli, Daae, & Eilertsen, 1995; Tucker, Bauer, Wagner, Harlam, & Sher, 1987; Waldinger & Gunderson, 1984; Wilberg et al., 1998). These studies with varied sample sizes speak to the relative efficacy of various forms of psychodynamic therapy but had too little in common in terms of treatment protocols to permit conclusions concerning the effectiveness of this approach. An Australian uncontrolled trial stands out in terms of methodological rigour (Meares, Stevenson, & Comerford, 1999; Stevenson & Meares, 1992; 1999). In this open trial 48 patients received twice weekly interpersonal self-psychological psychodynamic outpatient therapy over 12 months. The contrast was with patients on a waiting list for 12 months. Unfortunately allocation was not random and severity in the waiting list group was slightly less. 30% of the treatment group no longer met criteria for BPD at the end of one year. There was little indication of change in the control group. However intent to treat calculations would only estimate a 19% remission rate, which is comparable to the spontaneous change in follow-along studies. A waiting list control group is problematic and sometimes referred to as a nocebo group, as the implicit contingencies of being on a waiting list imply no change. A further large scale uncontrolled trial of psychodynamic psychotherapy is worth singling out. Dolan and colleagues (Dolan, Warren, & Norton, 1997) reported on the outcome of a therapeutic community run on strictly democratic principles, for example patients having a veto on the appropriateness of an admission. Of 598 patients referred, 239 were admitted and 137 (23%) returned assessment questionnaires at the one-year follow-up. About equal numbers of admitted and non-admitted patients returned the questionnaires, about 80% of whom met diagnostic criteria for BPD. Clinically significant change on self-reported borderline symptomatology was seen in 43% of the treated and 18% of the untreated patients (30 versus 12) (RRSTPPVSCONTROL=2.39, 95% CI: 1.34, 4.27). Length of stay was associated with improvement. The comparison group places profound limitations on the study, not just because of the absence of randomisation and the varied reasons for being in the no treatment group, but also because the pre-post time period covered in the treatment group was significantly longer (19 versus 12 months). Nevertheless, the study provides data concerning the likely change to be observed in a specialist but routine service context. The Cornell group (Clarkin et al., 2001) reported the outcomes of 23 female patients treated in transference focused psychotherapy. The trial which was a pilot for the Personality Disorders Institute Borderline Personality Disorder Research Foundation RCT (Clarkin et al.,

158

2004b) was a carefully conducted follow-along study of 23 female patients. After one year of treatment suicidal behaviour substantially decreased and the pre-post comparison of inpatient days suggested significant cost savings. Gabbard and colleagues (Gabbard et al., 2000) reported a prospective, naturalistic and uncontrolled study of consecutive patients admitted to the Menninger Hospital. Only 35% of the 216 completed in the sample were diagnosed with BPD. About half the patients had mixed PD or PD Not Otherwise Specified. An important feature of the study was the telephone follow-up at one year. GAF scores increased: only 3.7% had GAF scores above 50 on admission, which increased to 55% at discharge and 66% at follow-up. Other measures reflected a similar pattern. The study suggests that inpatient treatment can initiate improvement even in relatively severely dysfunctional patients. But the absence of a comparison group and the unknown selection bias introduced by limited participation reduces the generalisability of the data. Further the treatment package offered, whilst relatively consistent across patients, was not monitored in relation to each discipline. Given the wide diversity of length of stay it is hard to link progress to psychotherapeutic experience. Chiesa and colleagues (Chiesa & Fonagy, 2000, 2003; Chiesa, Fonagy, Holmes, & Drahorad, 2004; Chiesa, Fonagy, Holmes, Drahorad, & Harrison-Hall, 2002) reported a further controlled but not randomised trial of inpatient psychodynamic treatment. Two forms of hospitalbased treatment were contrasted with a general community based psychiatric treatment model. In the first protocol patients were admitted for approximately 12 months with no aftercare. In the second patients were admitted for only six months but this was followed by 12 months of outpatient therapy with community support. The third arm received community psychiatric care (medication and brief hospital admissions as necessary). 210 patients with at least one diagnosis of personality disorder were allocated according to geographical criteria into the three groups. Outcome was evaluated at 6, 12 and 24 months on self-harm and suicide attempts, and self reports of symptom severity and social adaptation. At 24 months only the phased or step-down condition showed improvements, while patients in the long-term residential model showed no improvements in self-harm, attempted suicide and number of readmissions. There were significant reductions in symptom severity, improvements in social adaptation and global functioning. Patients in the general psychiatric group showed no improvement in these variables, except for self-harm. 47% of the inpatient group and 73% and 71% of the step-down and general psychiatric group respectively reported no self-harm in the previous twelve months (RRINPVSTAU@24MONTHS=.66, 95% CI: . 47, .93; RRSDPVSTAU@24MONTHS=1.03, 95% CI: .81, 1.32). At 24 months more of the inpatient group had hospital admissions in the previous 12 months (49% for inpatient group compared to 11% for step-down and 33% for the general psychiatric care group; RRINPVSTAU@24MONTHS=1.5, 95% CI: .92, 2.45; RRSDPVSTAU@24MONTHS=.34, 95% CI: .14, .85). Thus in terms of clinical outcome the general psychiatric treatment group were somewhat inferior to the step-down group and superior to the inpatient group. The findings indicate that long-term inpatient therapy may be iatrogenic and may undermine some of the effective components of a treatment mode which results in substantial positive outcomes in more moderate doses. Only about 10-12% of the general psychiatric group showed clinically significant change in symptomatology and social adjustment compared to over half of the step-down group and about a quarter of the inpatient group. Mean Global Security Index (GSI) scores were more or less unchanged in the general psychiatry group. They were reduced by half a standard deviation in the inpatient group and by a whole standard deviation in the step-down group. Only one randomised controlled study is available. Bateman and Fonagy (Bateman & Fonagy, 1999; 2001; 2003) reported on a study of 38 patients assigned to specialist partial

159

hospitalisation or to routine care. Over 18 months partial hospitalisation showed significant gains over controls on measures of suicidality, self-harm and inpatient stay. These became apparent at 612 months of treatment and increased over time. Follow-up at 18 months, which included an intent to treat analysis, demonstrated that not only did patients in the programme maintain their gains but further improvements were observed. At the end of treatment 84% of the treatment as usual and 36% of the partial hospital patients had showed self-harming behaviours in the previous six months (RRSH@18MOS=.43, 95% CI: .24, .75). At 36 months 58% of the controls and 8% of the partial hospital patients had harmed themselves in the previous six months (RRSH@36MONTHS=.14, 95% CI: . 03, .55). A cost-benefit analysis suggested that in the course of the treatment additional costs of the programme are offset by reductions in inpatient and emergency room care costs as well as in reduced medication. The difference in costs per patient became apparent in the follow-up period. The mean annual cost of service utilisation was $15,500 for the TAU group and $3,200 for the partial hospitalisation group. This is an enormous discrepancy. The second controlled trial carried out by Clarkin, Kernberg, Levy and colleagues (Clarkin, Levy, Lenzenweger, & Kernberg, 2004a) is the most ambitious and comprehensive trial of psychodynamic psychotherapy in any context. It contrasts transference -focused psychotherapy (TFP) (Clarkin, Kernberg, & Yeomans, 1999) with dialectical behaviour therapy (Linehan, 1993) and psychoanalytic supportive psychotherapy (Rockland, 1987). TFP is based on Kernbergs object relations model and is a twice weekly outpatient therapy using clarification, confrontation, and transference interpretation. Psychodynamic supportive psychotherapy eschews transference interpretation, focuses on the strengthening of adaptive defences, forming an alliance and providing reassurance. All therapists were experienced in the conduct of their modality. Of 207 patients interviewed for the trial, 109 met the criteria. 19 refused randomisation but the remaining 90 were randomised to TFP, DBT or SPT. The baseline GAF score was about 50, quite severe for an outpatient sample. Results are available to 12 months. In all therapies GAF scores increased by about 10 points. BDI (Beck Depression Index) scores decreased significantly and social adjustment scores increased. There was no significant change in anxiety scores. The majority of patients showed improvement in their suicidality. Only a minority appeared to be getting worse. Hierarchical linear modelling showed that TFP and DBT improved significantly on suicidality but that patients in SPT treatment did not, and that all three treatment groups improved significantly in terms of global functioning and depression. On the Adult Attachment Interview (AAI) (Main & Goldwyn, 1998) ratings of coherence (closely related to attachment security) improved for all three groups. The improvement was most marked for the TFP group but this difference was not statistically significant. Reflective function scores (Fonagy, Target, Steele, & Steele, 1998), based on the AAI and related to mentalisation, showed slight improvements in the other two treatments but was only significant for the transference focused psychotherapy group (Levy & Clarkin, in press). Other than this significant interaction there were no differences between the treatment groups, except for a significantly higher early termination rate from DBT that could reflect more rapid improvement or lower acceptability of this treatment within this group. Antisocial Personality Disorder (ASPD) There are no trials of psychodynamic treatment of antisocial PD and only a small number of observational studies of individuals detained in high security settings (e.g. Reiss, Grubin, & Meux, 1996)Reiss, 1996 #wwfw2]. It is likely that at least some of these individuals would meet criteria for ASPD but this cannot be assumed. The studies are reviewed by Warren and colleagues

160

(Warren et al., 2003) . In general improvements are noted but the methodology is too weak to permit generalisation. More recently, Saunders (Saunders, 1996) contrasted CBT and STPP. The treatment was offered to men who were violent with their partners and of the 136 participants 40% met criteria for ASPD. No differences are reported between the groups in terms of recidivism and in the absence of a no treatment control group it is difficult to judge if either treatment was effective. A prison service in the UK (Grendon) is currently run along relatively coherent psychodynamic principles. Taylor (Taylor, 2000) describes outcomes from a seven year follow-up of 700 individuals who participated in this therapeutic community. The comparison groups in the report consist of demographically matched individuals who were never admitted to Grendon from the wait list group and 1,400 individuals treated from a general prison population. Attendance at the psychodynamic therapeutic community at Grendon was associated with a reduced rate of reoffending. Further there was a link between length of stay at Grendon and outcome. However when prior criminal histories are controlled for, the apparent impact of Grendon is reduced. Thornton and colleagues (Thornton, Mann , Bowers, & al, 1996) looked at a sub-group who were sex offenders. When matched with a group with similar forensic histories, those at the chronic end of the severity spectrum (at least two previous convictions for sex offences) had better outcomes. Cluster C (Anxious-Fearful) Personality Disorder Cluster-C personality disorders include avoidant personality disorder (social discomfort, timidity), dependent personality disorder (dependent on reassurance), and obsessive-compulsive personality disorder. These are the most prevalent personality disorders in the general population (10%) (Torgersen, Kringlen, & Cramer, 2001). We know of only one open trial of psychodynamic therapy that explicitly focuses on avoidant PD (Barber, Morse, Krakauer, Chittams, & Crits-Christoph, 1997). They used supportive expressive psychotherapy in the treatment of 38 individuals, 2/3 with avoidant and 1/3 with obsessive compulsive PD. Attrition was high, with 50% of the avoidant PDs leaving therapy prematurely. 40% of those with avoidant PD who stayed with therapy retained their diagnosis. Those with obsessive-compulsive PD had better retention rates and better outcomes. A small Norwegian trial (Svartberg, Stiles, & Seltzer, 2004) compared STPP with cognitive therapy for outpatients with cluster C personality disorder. 51 patients were randomly allocated to receive 40 weekly sessions of dynamic therapy (Malans approach) or cognitive therapy (Becks approach). Sessions were videotaped and adherence and integrity checks were performed with both therapies. Only two patients did not attend follow-up assessments at 6, 12 and 24 months. Both groups improved and continued to improve after treatment both symptomatologically and in terms of personality profile (Millons Clinical Multiaxial Inventory). On the SCL-90 38% and 17% respectively for STPP and Cognitive therapy were recovered, with asymptomatic status (below the clinical cutpoint) at the end of treatment (RRSCL90@12MONTHS=2.6, 95% CI: .94, 7.22). But at 2-year follow-up these had increased to 54% and 42% respectively (RRSCL90@24MONTHS=1.46, 95% CI: .8, 2.65). These figures were somewhat lower for personality change on the Millon (63% of the STPP group and 48% of the CBT group changed on this measure; RRMILLON@24MONTHS=1.39, 95% CI: .83, 2.31). The results clinical equivalence test (within interval of 20% of a zero difference) suggests that the group differences on the Millon are probably trivial but the differences on the SCL-90 may be of clinical significance. Sadly, the study is underpowered to detect more than a large effect size which is unlikely to be observed in this kind of context.

161

A randomised trial compared STPP for predominantly cluster C personality disordered individuals along the lines developed by Malan and Davanloo (n=31) with brief adaptive psychotherapy (BAP) developed by the authors (Winston et al., 1994) (n=32). There was also a waiting list control group (n=26). The former form of STPP is believed by the authors to be more confrontational but both appear to address defensive behaviour and elicit affect in interpersonal contexts. 32 patients were randomly allocated to BAP and 31 to more traditional STPP. 25 completed STPP and 30 BAP. Mean treatment length was 40 sessions but the waiting list control group lasted only 15 weeks. A large number of therapists participated. Treatment manuals and videotaping were employed for adherence checks. A variable length follow-up reached about two thirds of the treated group. The two treated groups both showed significant change on the GSI of the SCL-90 of approximately one standard deviation and some change on the social adjustment scale. There were no significant differences between the two treated groups - not surprising given the similarity of the approaches. There were some differences between the groups in terms of gender and Axis I diagnoses and heterogeneity as to Axis II diagnoses. The study was underpowered to look at specific benefits of each therapy in relation to particular PD types. An earlier study by the same group (Winston et al., 1991) contrasted contrasting the same therapies, reported essentially the same results with similar effect sizes on the GSI and the Social Adjustment Scale (SAS). This study had aA more robust follow-up of this study at 18 months (Winston et al., 1994) indicated that the gains were maintained.

LONG TERM PSYCHOTHERAPY

In the previous sections we considered evidence available to support therapeutic interventions which are derivatives of psychoanalysis. However, there is a certain degree of disingenuity in psychoanalysis embracing these investigations. Most analysts would consider that the aims and methods of short-term once a week psychotherapy are not comparable to full analysis. What do we know about the value of intensive and long term psychodynamic treatment? Here the evidence base becomes even more patchy and we cannot restrict the review to randomized controlled trials. The Boston Psychotherapy Study (Stanton et al., 1984) compared long term psychoanalytic therapy (two or more times a week) with supportive therapy for clients with schizophrenia in a randomized controlled design. On the whole clients who received psychoanalytic therapy fared no better than those who received supportive treatment. The partial-hospital RCT (Bateman & Fonagy, 1999), included in the psychoanalytic arm of the treatment, included therapy groups three times a week as well as individual therapy once or twice a week over an 18 month period. A further controlled trial of intensive psychoanalytic treatment of children with chronically poorly controlled diabetes reported significant gains in diabetic control in the treated group which was maintained at one year follow-up (Moran, Fonagy, Kurtz, Bolton, & Brook, 1991). Experimental single case studies carried out with the same population supported the causal relationship between interpretive work and improvement in diabetic control and physical growth (Fonagy & Moran, 1991). The work of Chris Heinicke also suggests that four or five times weekly sessions may generate more marked improvements in children with specific learning difficulties than a less intensive psychoanalytic intervention (Heinicke & Ramsey-Klee, 1986). One of the most interesting studies to emerge recently was the Stockholm Outcome of Psychotherapy and Psychoanalysis Project (Blomberg, Lazar, & Sandell, 2001; Grant & Sandell, 2004; Sandell et al., 2000). The study followed 756 persons who received national insurance funded treatment for up to three years in psychoanalysis or psychoanalytic psychotherapy. The

162

groups were matched on many clinical variables. Four or five times weekly analysis had similar outcomes at termination when compared with one to two sessions per week psychotherapy. During the follow-up period, psychotherapy patients did not change but those who had had psychoanalysis continued to improve, almost to a point where their scores were indistinguishable from those obtained from a non-clinical Swedish sample. While the results of the study are positive for psychoanalysis, certain findings are quite challenging. For example, therapists whose attitude to clinical process most closely resembled that of a so-called classical analyst (neutrality, exclusive orientation to insight) had psychotherapy clients with the worst, commonly negative, outcomes. The German Psychoanalytic Association undertook a major follow-up study of psychoanalytic treatments undertaken in that country between 1990 and 1993 (LeuzingerBohleber, Stuhr, Ruger, & Beutel, 2003; Leuzinger-Bohleber & Target, 2002). A representative sample (n = 401) of all the patients who had terminated their psychoanalytic treatments with members of the German Psychoanalytical Association (DPV) were followed up. Between 70 and 80 per cent of the patients achieved (average 6.5 years after the end of treatment) good and stable psychic changes according to the evaluations of the patients themselves, their analysts, independent psychoanalytic and non-psychoanalytic experts, and questionnaires commonly applied in psychotherapy research. The evaluation of mental health costs showed a cost reduction through fewer days of sick leave during the seven years following the end of long-term psychoanalytic treatments. Qualitative analysis of the data also pointed to the value that patients continued to attach to their analytic experience. In the absence of pre-treatment measures it is impossible to estimate the size of the treatment effect. Another large pre-post study of psychoanalytic treatments has examined the clinical records of 763 children who were evaluated and treated at the Anna Freud Centre, under the close supervision of Anna Freud (Fonagy & Target, 1994, 1996; Target & Fonagy, 1994a, 1994b). Children with certain disorders (e.g. depression, autism, conduct disorder) appeared to benefit only marginally from psychoanalysis or psychoanalytic psychotherapy. Interestingly, children with severe emotional disorders (three or more Axis I diagnoses) did surprisingly well in psychoanalysis, although they did poorly in once or twice a week psychoanalytic psychotherapy. Younger children derived greatest benefit from intensive treatment. Adolescents appeared not to benefit from the increased frequency of sessions. The importance of the study is perhaps less in demonstrating that psychoanalysis is effective, although some of the effects on very severely disturbed children were quite remarkable, but more in identifying groups for whom the additional effort involved in intensive treatment appeared not to be warranted. The Research Committee of the International Psychoanalytic Association has recently prepared a comprehensive review of North American and European outcome studies of psychoanalytic treatment (Fonagy, Kachele et al., 2002). The committee concluded that existing studies failed to unequivocally demonstrate that psychoanalysis is efficacious relative to either an alternative treatment or an active placebo. A range of methodological and design problems was identified including absence of intent to treat controls, heterogeneous patient groups, lack of random assignments, the failure to use independently administered standardized measures of outcome, etc. Nevertheless, the report, which ran to several hundred pages and briefly describes more than fifty studies, is encouraging to psychoanalysts. Another overview (Gabbard, Gunderson, & Fonagy, 2002) suggested that psychoanalytic treatments may be necessary when other treatments prove to be ineffective. The authors concluded that psychoanalysis appears to be consistently helpful to patients with milder disorders and somewhat helpful to those with more severe disturbances. More controlled studies are necessary to confirm these impressions. A number

163

of studies testing psychoanalysis with state of the art methodology are ongoing and are likely to produce more compelling evidence over the next years. Despite the limitations of the completed studies, evidence across a significant number of pre-post investigations suggests that psychoanalysis appears to be consistently helpful to patients with milder (neurotic) disorders and somewhat less consistently so for other, more severe, groups. Across a range of uncontrolled or poorly controlled cohort studies, mostly carried out in Europe, longer intensive treatments tended to have better outcomes than shorter, non-intensive treatments (demonstration of a dose-effect relationship). The impact of psychoanalysis was apparent beyond symptomatology, in measures of work functioning and reductions in health care costs.

CONCLUSIONS
Considerable evidence has accumulated for the evidence of psychoanalytic approaches for a range of diagnostic conditions. The strength of the evidence varies across clinical groups and in the case of some groups there is little current evidence supporting the approach. In no area is the evidence compelling but in most areas where systematic investigation has been carried out, outcomes are comparable to those obtained by other therapeutic methods. There are disorders where the outcome of psychoanalytic therapy is in certain respects better than alternative treatments (e.g. borderline personality disorder). In other areas (e.g. depression) there are opportunities for the psychodynamic approach created by the known limitations of rival orientations. In relation to treatment resistant conditions, technical innovation is called for and this has been a significant barrier to psychoanalytically oriented investigations. Even where original and effective approaches have been developed, replication of findings by groups other than those responsible for the development of the program are hard to find. There can be few valid excuses for the currently thin evidence base of psychoanalytic treatment. In the same breath that we, as psychoanalysts, often claim to be at the intellectual origin of other talking cures (e.g. systemic therapy, cognitive behaviour therapy), we also seek to shelter behind the relative immaturity of the discipline as an account for the absence of research evidence for its efficacy. Yet the evidence base of these derivatives of psychoanalytic therapy has been far more firmly established than evidence for the approach at the root of the psychotherapy movement (Holmes, 2002). Of course there are reasons for this reasons such as the long term nature of the therapy, the subtlety and complexity of its procedures, the elusiveness of its self-declared outcome goals, and the incompatibility of direct observation and the need for absolute confidentiality. None of these reasons can stand up to careful scrutiny. The scientific method has been, in recent years, extended to a wide range of highly complex phenomena, and our understanding of the psychological processes that are at work in mental disorders and their psychosocial therapy is a focus of intense inquiry from a range of perspectives. Neither the subject matter nor the difficulties inherent to its systematic investigation account for the relative lack of support for empirical and systematic qualitative studies of the process and outcome of psychoanalytic therapies. Currently, political considerations make funded research in this field difficult but historically this has not always been the case. A more likely reason for the absence of psychoanalytic outcome research lies in the fundamental incompatibilities in the world view espoused by psychoanalysis and most of current social science (Whittle, 2000) that will require a shift in epistemology on the part of psychoanalytic psychotherapists. There are several components to this attitude change: (1) the incorporation of data gathering methods beyond the anecdotal, methods that are now widely available in social and

164

biological science; (2) moving psychoanalytic constructs from the global to the specific, which will facilitate cumulative data gathering and identifying the psychological mechanisms which account for change in psychodynamic therapy; (3) routine consideration of alternative accounts for behavioural observations of change; (4) increasing psychoanalytic sophistication concerning social and contextual influences on pathological behaviour and its response to treatment; (5) ending the splendid isolation of psychoanalysis by undertaking active collaboration with other scientific and clinical disciplines; (6) using the knowledge-base of psychoanalysis to generate innovative treatment approaches to currently treatment resistant conditions; (7) integrating successful psychotherapeutic manipulations from other disciplines into a psychodynamic approach; (8) identifying clinical groups for whom the psychodynamic method is particularly effective and (9) adopting a scientific attitude that celebrates the value of the replication of observations rather than their uniqueness. Rather than fearing that fields adjacent to psychoanalysis might destroy the unique insights offered by long term intensive individual therapy, psychoanalysts must embrace the rapidly evolving knowledge chain, focused at different levels of the study of brain-behaviour relationships. As Kandel (1998; 1999) has pointed out, this may be the major route to the preservation of the hard-won insights of psychoanalysis.

165

REFERENCES
Ablon, J. S., & Jones, E. E. (1998). How expert clinicians' prototypes of an ideal treatment correlate with outcome in psychodynamic and cognitive-behavioral therapy. Psychotherapy Research, 8, 71-83. Ablon, J. S., & Jones, E. E. (1999). Psychotherapy process in the National Institute of Mental Health Treatment of Depression Collaborative Research Program. Journal of Consulting & Clinical Psychology, 67(1), 64-75. Ablon, J. S., & Jones, E. E. (2002). Validity of controlled cinical trials of psychotherapy: findings from the NIMH Treatment of Depression Collaborative Research Program. Am J Psychiatry, 159(5), 775-783. Anon. (1992). Cross design synthesis: A new strategy for studying medical outcomes. Lancet, 340, 944-946. Antikainen, R., Hintikka, J., Lehtonen, J., Koponen, H., & Arstila, A. (1995). A prospective threeyear follow-up study of borderline personality disorder inpatients. Acta Psychiatr Scand, 92(5), 327-335. Bachar, E., Latzer, Y., Kreitler, S., & Berry, E. M. (1999). Empirical comparison of two psychological therapies. Self psychology and cognitive orientation in the treatment of anorexia and bulimia. J Psychother Pract Res, 8(2), 115-128. Barber, J. P., Morse, J. Q., Krakauer, I. D., Chittams, J., & Crits-Christoph, K. (1997). Change in obsessive-compulsive and avoidant personality disorders following time-limited supportive-expressive therapy. Psychotherapy, 34(2), 133-143. Barkham, M., Rees, A., Shapiro, D. A., Stiles, W. B., Agnew, R. M., Halstead, J., et al. (1996). Outcomes of time-limited psychotherapy in applied settings: replicating the Second Sheffield Psychotherapy Project. Journal of Consulting & Clinical Psychology, 64(5), 1079-1085. Barkham, M., Rees, A., Stiles, W. B., Shapiro, D. A., Hardy, G. E., & Reynolds, S. (1996). Doseeffect relations in time-limited psychotherapy for depression. Journal of Consulting & Clinical Psychology, 64(5), 927-935. Barlow, D. H., Gorman, J. M., Shear, M. K., & Woods, S. W. (2000). Cognitive-behavioral therapy, imipramine, or their combination for panic disorder: A randomized controlled trial. Jama, 283(19), 2529-2536. Bateman, A., & Fonagy, P. (1999). The effectiveness of partial hospitalization in the treatment of borderline personality disorder - a randomised controlled trial. American Journal of Psychiatry, 156, 1563-1569. Bateman, A., & Fonagy, P. (2001). Treatment of borderline personality disorder with psychoanalytically oriented partial hospitalization: an 18-month follow-up. American Journal of Psychiatry, 158(1), 36-42. Bateman, A., & Fonagy, P. (2003). Health service utilization costs for borderline personality disorder patients treated with psychoanalytically oriented partial hospitalization versus general psychiatric care. Am J Psychiatry, 160(1), 169-171. Bateman, A. W., & Fonagy, P. (2004). Psychotherapy for Borderline Personality Disorder: Mentalization Based Treatment. Oxford: Oxford University Press. Beck, A., Sokol, L., Clark, D., Berchick, R., & Wright, F. (1992). A crossover study of focused cognitive therapy for panic disorder. American Journal of Psychiatry, 149, 778-783.

166

Bellack, A., Hersen, M., & Himmelhoch, J. (1981). Social skills training compared with pharmacotherapy and psychotherapy for depression. American Journal of Psychiatry, 138, 1562-1567. Beutel, M., Thiede, R., Wiltlink, J., & Sobez, I. (2001). Effectiveness of behavioral and psychodynamic in-patient treatment of severe obesity - first results from a randomized study. International Journal of Obesity, 25, 96-98. Blomberg, J., Lazar, A., & Sandell, R. (2001). Outcome of patients in long-term psychoanalytical treatments. First findings of the Stockholm Outcome of Psychotherapy and Psychoanalysis (STOPP) study. Psychotherapy Research, 11, 361-382. Blum, N., Pfohl, B., John, D. S., Monahan, P., & Black, D. W. (2002). STEPPS: a cognitivebehavioral systems-based group treatment for outpatients with borderline personality disorder--a preliminary report. Compr Psychiatry, 43(4), 301-310. Borkovec, T. D., & Ruscio, A. M. (2001). Psychotherapy for generalized anxiety disorder. J Clin Psychiatry, 62(Suppl 11), 37-42; discussion 43-35. Brandsma, J., Maultsby, M., & Welsh, R. (1980). The outpatient treatment of alcoholism: A review and a comparative study. Baltimore: University Park Press. Brewer, D. D., Catalano, R. F., Haggerty, K., Gainey, R. R., & Fleming, C. B. (1998). A metaanalysis of predictors of continued drug use during and after treatment for opiate addiction. Addiction, 93(1), 73-92. Brom, D., Kleber, R., & Defares, P. (1989). Brief psychotherapy for PTSD. Journal of Consulting and Clinical Psychology, 57, 607-612. Brown, T. A., Campbell, L. A., Lehman, C. L., Grisham, J. R., & Mancill, R. B. (2001). Current and lifetime comorbidity of the DSM-IV anxiety and mood disorders in a large clinical sample. J Abnorm Psychol, 110(4), 585-599. Burnand, Y., Andreoli, A., Kolatte, E., Venturini, A., & Rosset, N. (2002). Psychodynamic psychotherapy and clomipramine in the treatment of major depression. Psychiatr Serv, 53(5), 585-590. Busch, F., Milrod, B., Cooper, A., & Shapiro, T. (1996). Grand Rounds: Panic Focused Psychodynamic Psychotherapy. Journal Of Psychotherapy Research and Practice., 5, 7283. Busch, F., Milrod, B., & Singer, M. (2000). Theory and technique in the psychodynamic treatment of panic disorder. Review Theories of Psychiatry, 4100, 16-17. Chambless, D. L., & Hollon, S. (1998). Defining empirically supported therapies. Journal of Consulting and Clinical Psychology, 66, 7-18. Chambless, D. L., Sanderson, W. C., Shoham, V., Johnson, S. B., Pope, K. S., Crits-Christoph, P., et al. (1996). An update on clinically validated therapies. The Clinical Psychologist, 49, 518. Chiesa, M., & Fonagy, P. (2000). Cassel Personality Disorder Study. Methodology and treatment effects. Br J Psychiatry, 176, 485-491. Chiesa, M., & Fonagy, P. (2003). Psychosocial treatment for severe personality disorder. 36-month follow-up. Br J Psychiatry, 183, 356-362. Chiesa, M., Fonagy, P., Holmes, J., & Drahorad, C. (2004). Residential versus community treatment of personality disorder: A comparative study of three treatment programs. American Journal of Psychiatry.

167

Chiesa, M., Fonagy, P., Holmes, J., Drahorad, C., & Harrison-Hall, A. (2002). Health service use costs by personality disorder following specialist and nonspecialist treatment: a comparative study. J Personal Disord, 16(2), 160-173. Chilvers, R., Harrison, G., Sipos, A., & Barley, M. (2002). Evidence into practice: Application of psychological models of change in evidence-based implementation. Br J Psychiatry, 181(2), 99-101. Churchill, R., Hunot, V., Corney, R., Knapp, M., McGuire, H., Tylee, A., et al. (2001). A systematic review of controlled trials of the effectiveness and cost-effectiveness of brief psychological treatments for depression. Health Technology Assessment, 5(35). Clarke, M., & Oxman, A. D. (1999). Cochrane Reviewers' Handbook 4.0 [updated July 1999]. In Review Manager [Computer Program]: Version 4.0. Oxford: The Cochrane Collaboration. Clarkin, J. F., Foelsch, P. A., Levy, K. N., Hull, J. W., Delaney, J. C., & Kernberg, O. F. (2001). The development of a psychodynamic treatment for patients with borderline personality disorder: a preliminary study of behavioral change. J Personal Disord, 15(6), 487-495. Clarkin, J. F., Kernberg, O. F., & Yeomans, F. (1999). Transference-Focused Psychotherapy for Borderline Personality Disorder Patients. New York, NY: Guilford Press. Clarkin, J. F., Levy, K. N., Lenzenweger, M. F., & Kernberg, O. F. (2004a). The Personality Disorders Institute/Borderline Personality Disorder Research Foundation randomized control trial for borderline personality disorder: Progress report. Paper presented at the Annual Meeting of the Society for Psychotherapy Research, Rome, Italy. Clarkin, J. F., Levy, K. N., Lenzenweger, M. F., & Kernberg, O. F. (2004b). The Personality Disorders Institute/Borderline Personality Disorder Research Foundation randomized control trial for borderline personality disorder: rationale, methods, and patient characteristics. J Personal Disord, 18(1), 52-72. Connolly, M. B., Crits-Christoph, P., Shappell, S., Barber, J. P., Luborsky, L., & Shaffer, C. (1999). Relation of transference interpretations to outcome in the early sessions of brief supportive-expressive psychotherapy. Psychotherapy Research, 9, 485-495. Covi, L., & Lipman, R. S. (1987). Cognitive-behavioral group psychotherapy combined with imipramine in major depression. Psychopharmacology Bulletin, 23, 173-176. Craske, M. G., Maidenberg, E., & Bystritsky, A. (1995). Brief cognitive-behavioral versus nondirective therapy for panic disorder. J Behav Ther Exp Psychiatry, 26(2), 113-120. Crisp, A. H., Norton, K., Gowers, S., Halek, C., Bowyer, C., Yeldham, D., et al. (1991). A controlled study of the effect of therapies aimed at adolescent and family psychopathology in anorexia nervosa. British Journal of Psychiatry, 159, 325-333. Crits-Christoph, P. (1992). The efficacy of brief dynamic psychotherapy: A meta-analysis. American Journal of Psychiatry, 159, 325-333. Crits-Christoph, P., Connolly, M. B., Azarian, K., Crits-Christoph, K., & Shappell, S. (1996). An open trial of brief supportive-expressive psychotherapy in the treatment of generalized anxiety disorder. Psychotherapy, 33, 418-430. Crits-Christoph, P., Siqueland, L., Blaine, J., Frank, A., Luborsky, L., Onken, L. S., et al. (1997). The National Institute on Drug Abuse Collaborative Cocaine Treatment Study. Rationale and methods. Arch Gen Psychiatry, 54(8), 721-726. Crits-Christoph, P., Siqueland, L., Blaine, J., Frank, A., Luborsky, L., Onken, L. S., et al. (1999). Psychosocial treatments for cocaine dependence: National Institute on Drug Abuse Collaborative Cocaine Treatment Study. Arch Gen Psychiatry, 56(6), 493-502.

168

Crits-Christoph, P., Siqueland, L., McCalmont, E., Weiss, R. D., Gastfriend, D. R., Frank, A., et al. (2001). Impact of psychosocial treatments on associated problems of cocaine-dependent patients. Journal of Consulting & Clinical Psychology, 69(5), 825-830. Crumbach, J. C., & Carr, G. L. (1979). Treatment of alcoholics with logotherapy. International Journal of the Addictions, 14, 847-853. Cryer, L., & Beutler, L. (1980). Group therapy: An alternative approach for rape victims. Journal of Sex and Marital Therapy, 6, 40-46. Dare, C., & Eisler, E. (1997). Family therapy for anorexia nervosa. In D. G. Garner & P. E. Garfinkel (Eds.), Handbook of Treatment for Eating Disorders (2 ed.). New York: Guilford. Dare, C., Eisler, I., Russell, G., Treasure, J., & Dodge, L. (2001). Psychological therapies for adults with anorexia nervosa: randomised controlled trial of out-patient treatments. Br J Psychiatry, 178, 216-221. Dawe, S., Powell, J., Richards, D., Gossop, M., Marks, I., Strang, J., et al. (1993). Does postwithdrawal cue exposure improve outcome in opiate addiction? A controlled trial. Addiction, 88(9), 1233-1245. de Jonghe, F., Kool, S., van Aalst, G., Dekker, J., & Peen, J. (2001). Combining psychotherapy and antidepressants in the treatment of depression. J Affect Disord, 64(2-3), 217-229. de Jonghe, F., Rijnierse, P., & Janssen, R. (1994). Psychoanalytic supportive psychotherapy. Journal of the American Psychoanalytic Association, 42, 421446. Department of Health. (1995). Strategic Report on the Psychotherapies. London: HMSO. Dolan, B., Warren, F., & Norton, K. (1997). Change in borderline symptoms one year after therapeutic community treatment for severe personality disorder. Br J Psychiatry, 171, 274-279. Durham, R. C., Chambers, J. A., MacDonald, R. R., Power, K. G., & Major, K. (2003). Does cognitive-behavioural therapy influence the long-term outcome of generalized anxiety disorder? An 8-14 year follow-up of two clinical trials. Psychol Med, 33(3), 499-509. Durham, R. C., Fisher, P. L., Treliving, L. R., Hau, C. M., Richard, K., & Stewart, J. B. (1999). One year follow-up of cognitive therapy, analytic psychotherapy and anxiety management training for generalized anxiety disorder: symptom change, medication usage and attitudes to treatment. Behavioural and Cognitive Psychotherapy, 27, 19-35. Durham, R. C., Murphy, T., Allan, T., Richard, K., Treliving, L. R., & Fenton, G. W. (1994). Cognitive therapy, analytic psychotherapy and anxiety management training for generalized anxiety disorder. British Journal of Psychiatry, 165, 315-323. Eisler, I., Dare, C., Russell, G. F., Szmukler, G., le Grange, D., & Dodge, E. (1997). Family and individual therapy in anorexia nervosa. A 5-year follow-up. Arch Gen Psychiatry, 54(11), 1025-1030. Elkin, I. (1994). The NIMH treatment of depression collaborative research program: Where we began and where we are. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th ed.). New York: Wiley. Finney, J. W., & Monahan, S. C. (1996). The cost-effectiveness of treatment for alcoholism: a second approximation. J Stud Alcohol, 57(3), 229-243. Fisher, P. L., & Durham, R. C. (1999). Recovery rates in generalized anxiety disorder following psychological therapy: an analysis of clinically significant change in the STAI-T across outcome studies since 1990. Psychol Med, 29(6), 1425-1434.

169

Fonagy, P. (1999a). Process and outcome in mental health care delivery: a model approach to treatment evaluation. Bulletin Of The Menninger Clinic, 63(3), 288-304. Fonagy, P. (1999b). The relation of theory and practice in psychodynamic therapy. Journal of Clinical Child Psychology, 28, 513-520. Fonagy, P. (2003). Some complexities in the relationship of psychoanalytic theory to technique. Psychoanalytic Quarterly, 72(1), 13-47. Fonagy, P., Kachele, H., Krause, R., Jones, E., Perron, R., Clarkin, J., et al. (2002). An open door review of outcome studies in psychoanalysis (2nd ed.). London: International Psychoanalytical Association. Fonagy, P., & Moran, G. S. (1991). Studies of the efficacy of child psychoanalysis. Journal of Consulting and Clinical Psychology, 58, 684-695. Fonagy, P., & Target, M. (1994). The efficacy of psychoanalysis for children with disruptive disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 4555. Fonagy, P., & Target, M. (1996). Predictors of outcome in child psychoanalysis: A retrospective study of 763 cases at the Anna Freud Centre. Journal of the American Psychoanalytic Association, 44, 27-77. Fonagy, P., Target, M., Cottrell, D., Phillips, J., & Kurtz, Z. (2002). What Works For Whom? A Critical Review of Treatments for Children and Adolescents. New York: Guilford. Fonagy, P., Target, M., Steele, H., & Steele, M. (1998). Reflective-Functioning Manual, version 5.0, for Application to Adult Attachment Interviews. London: University College London. Gabbard, G. O., Coyne, L., Allen, J. G., Spohn, H., Colson, D. B., & Vary, M. (2000). Evaluation of intensive inpatient treatment of patients with severe personality disorders. Psychiatr Serv, 51(7), 893-898. Gabbard, G. O., Gunderson, J. G., & Fonagy, P. (2002). The place of psychoanalytic treatments within psychiatry. Archives of General Psychiatry, 59, 505-510. Gallagher-Thompson, D., & Steffen, A. M. (1994). Comparative effects of cognitive-behavioral and brief psychodynamic psychotherapies for depressed family caregivers. Journal of Consulting and Clinical Psychology, 62, 543-549. Garner, D. M., Rockert, W., Davis, R., & Garner, M. D. (1993). A comparison between CBT and supportive expressive therapy for bulimia nervosa. American Journal of Psychiatry, 150, 37-46. Gloaguen, V., Cottraux, J., Cucherat, M., & Blackburn, I. (1998). A meta-analysis of the effects of cognitive therapy in depressed patients. Journal of Affective Disorders, 49(1), 59-72. Goldfried, M. R., & Wolfe, B. E. (1996). Psychotherapy practice and research: Repairing a strained alliance. American Psychologist, 51, 1007-1016. Gowers, S., Norton, K., Halek, C., & Crisp, A. H. (1994). Outcome of outpatient psychotherapy in a random allocation treatment study of anorexia nervosa. International Journal of Eating Disorders, 15, 165-177. Grant, J., & Sandell, R. (2004). Close family or mere neighbours? Some empirical data on the differences between psychoanalysis and psychotherapy. In P. Richardson, H. Kchele & C. Renlund (Eds.), Research on psychoanalytic psychotherapy with adults (pp. 81-108). London: Karnac. Guthrie, E., Moorey, J., Margison, F., Barker, H., Palmer, S., McGrath, G., Tomenson, B., Creed F. (1999). Cost-effectiveness of brief psychodynamic-interpersonal therapy in high utilizers of psychiatric services. Archives of General Psychiatry, 56(6), 519-526.

170

Hall, A., & Crisp, A. N. (1987). Brief psychotherapy in the treatment of anorexia nervosa. British Journal of Psychiatry, 151, 185-191. Hay, P. J., & Bacaltchuk, J. (2001). Psychotherapy for bulimia nervosa and binging (Cochrane Review). In The Cochrane Library, Issue 4, 2002. Oxford: Update Software. Hayes, A. M., Castonguay, L. G., & Goldfried, M. R. (1996). Effectiveness of targeting the vulnerability factors of depression in cognitive therapy. Journal of Consulting & Clinical Psychology, 64(3), 623-627. Heinicke, C. M., & Ramsey-Klee, D. M. (1986). Outcome of child psychotherapy as a function of frequency of sessions. Journal of the American Academy of Child Psychiatry, 25, 247-253. Hersen, M., Bellack, A. S., Himmelhoch, J. M., & Thase, M. E. (1984). Effects of social skills training, amitriptyline and psychotherapy in unipolar depressed women. Behavior Therapy, 15, 21-40. Hickman, M., Cox, S., Harvey, J., Howes, S., Farrell, M., Frischer, M., et al. (1999). Estimating the prevalence of problem drug use in inner London: a discussion of three capturerecapture studies. Addiction, 94(11), 1653-1662. Higgitt, A., & Fonagy, P. (2002). Clinical effectiveness. Br J Psychiatry, 181(2), 170-174. Hoagwood, K., Hibbs, E., Brent, D., & Jensen, P. (1995). Introduction to the special section: Efficacy and effectiveness in studies of child and adolescent psychotherapy. Journal of Consulting and Clinical Psychology, 63, 683-687. Holmes, J. (2002). All you need is cognitive behaviour therapy? British Medical Journal, 324(7332), 288-290; discussion 290-284. Hopper, E. (1995). A Psychoanalytical Theory Of Drug Addiction: Unconscious Fantasies Of Homosexuality, Compulsions And Masturbation Within The Context Of Traumatogenic Processes. Int. J. Psycho-Anal., 76, 1121-1142. Horowitz, M. J., Marmar, C., Weiss, D. S., DeWitt, K. N., & Rosenbaum, R. (1984). Brief psychotherapy of bereavement reactions: The relationship of process to outcome. Archives of General Psychiatry, 41, 438-448. Howard, K., Lueger, R., Maling, M., & Martinovitch, Z. (1993). A phase model of psychotherapy: Causal mediation of outcome. Journal of Consulting & Clinical Psychology, 61, 678-685. Johnson, B. (1999). Three Perspectives On Addiction. J. Amer. Psychoanal. Assn., 47, 791-815. Kandel, E. R. (1998). A new intellectual framework for psychiatry. American Journal of Psychiatry, 155, 457-469. Kandel, E. R. (1999). Biology and the future of psychoanalysis: A new intellectual framework for psychiatry revisited. American Journal of Psychiatry, 156, 505-524. Kang, S. Y., Kleinman, P. H., Woody, G. E., Millman, R. B., Todd, T. C., Kemp, J., et al. (1991). Outcomes for cocaine abusers after once-a-week psychosocial therapy. Am J Psychiatry, 148(5), 630-635. Karterud, S., Vaglum, S., Friis, S., Irion, T., Johns, S., & Vaglum, P. (1992). Day hospital therapeutic community treatment for patients with personality disorders. An empirical evaluation of the containment function. J Nerv Ment Dis, 180(4), 238-243. Kasvikis, Y., Bradley, B., Powell, J., Marks, I., & Gray, J. A. (1991). Postwithdrawal exposure treatment to prevent relapse in opiate addicts: a pilot study. Int J Addict, 26(11), 11871195. Kazdin, A. E. (1998). Research design in clinical psychology. Needham Heights, MA: Allyn & Bacon.

171

Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., et al. (1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Archives of General Psychiatry, 51, 8-19. Kessler, R. C., Stang, P., Wittchen, H. U., Stein, M., & Walters, E. E. (1999). Lifetime comorbidities between social phobia and mood disorders in the US National Comorbidity Survey. Psychol Med, 29(3), 555-567. Kissin, B., & Gross, M. M. (1968). Drug therapy in alcoholism. American Journal of Psychiatry, 125, 31-41. Klein, D. F., Zitrin, C. M., Woerner, M. G., & Ross, D. C. (1983). Treatment of phobias. II. Behavior therapy and supportive psychotherapy: are there any specific ingredients? Arch Gen Psychiatry, 40(2), 139-145. Knekt, P., & Lindfors, O. (Eds.). (2004). A randomized trial of the effects of four forms of psychotherapy on depressive and anxiety disorders: Design methods and results on the effectiveness of short term spychodynamic psychotherapy and solution focused therapy during a 1-year follow-up (Vol. 77). Helsinki: Social Insurance Institution. Koons, C. R., Robins, C. J., Tweed, J. L., Lynch, T. R., Gonzalez, A. M., Morse, J. Q., et al. (2001). Efficacy of dialectical behavior therapy in women veterans with borderline personality disorder. Behavior Therapy, 32, 371-390. Kopta, S., Howard, K., Lowry, J., & Beutler, L. (1994). Patterns of symptomatic recovery in psychotherapy. Journal of Consulting and Clinical Psychology, 62, 1009-1016. Kornblith, S. J., Rehm, L. P., O'Hara, M. W., & Lamparski, D. M. (1983). The contribution of selfreinforcement training and behavioral assignments to the efficacy of self-control therapy for depression. Cognitive Therapy and Research, 7, 499-528. Kraus, L., Augustin, R., Frischer, M., Kummler, P., Uhl, A., & Wiessing, L. (2003). Estimating prevalence of problem drug use at national level in countries of the European Union and Norway. Addiction, 98(4), 471-485. Lambert, M.J., Okiishi, J.C., Finch, A.E., & Johnson,L.D. (1998). Outcome assessment: From conceptualization to implementation. Professional Psychology: Research and Practice 29: 63-70. Leff, J., Vearnals, S., Brewin, C. R., Wolff, G., Alexander, B., Asen, E., et al. (2000). The London Depression Intervention Trial. Randomised controlled trial of antidepressants v. couple therapy in the treatment and maintenance of people with depression living with a partner: clinical outcome and costs. Br J Psychiatry, 177, 95-100. Leichsenring, F. (2001). Comparative effects of short-term psychodynamic psychotherapy and cognitive-behavioral therapy in depression: a meta-analytic approach. Clin Psychol Rev, 21(3), 401-419. Leichsenring, F., & Leibing, E. (2003). The effectiveness of psychodynamic therapy and cognitive behavior therapy in the treatment of personality disorders: a meta-analysis. Am J Psychiatry, 160(7), 1223-1232. Leuzinger-Bohleber, M., Stuhr, U., Ruger, B., & Beutel, M. (2003). How to study the quality of psychoanalytic treatments and their long-term effects on patients' well-being: a representative, multi-perspective follow-up study. International Journal of Psychoanalysis, 84, 263-290. Leuzinger-Bohleber, M., & Target, M. (Eds.). (2002). The Outcomes of Psychoanalytic Treatment. London: Whurr.

172

Levinson, T., & Sereny, G. (1969). An experimental evaluation of insight therapy for the chronic alcoholic. Canadian Psychiatric Association Journal, 14, 143-146. Levy, K. N., & Clarkin, J. F. (in press). To examine the mechanisms of change in the the treatment of borderline personality disorder. Journal of Clinical Psychology. Linehan, M. M. (1993). Cognitive-Behavioural Treatment of Borderline Personality Disorder. New York: Guilford. Linehan, M. M., Schmidt, H., Dimeff, L. A., Craft, J. C., Kanter, J., & Comtois, K. A. (1999). Dialectical behavior therapy for patients with borderline personality disorder and drug dependence. American Journal on Addictions, 8, 279-292. Lipsitz, J. D., Markowitz, J. C., Cherry, S., & Fyer, A. J. (1999). Open trial of interpersonal psychotherapy for the treatment of social phobia. Am J Psychiatry, 156(11), 1814-1816. Luborsky, L. (1984). Principles of psychoanalytic psychotherapy: A manual for supportiveexpressive treatment. New York: Basic Books. Main, M., & Goldwyn, R. (1998). Adult attachment scoring and classification systems. Unpublished manuscript, University of California at Berkeley. Malan, D. H. (1976). Toward the Validation of Dynamic Psychotherapy. New York: Plenum Press. Marmar, C. R., Horowitz, M. J., Weiss, D. S., Wilner, N. R., & Kaltreider, N. B. (1988). A controlled trial of brief psychotherapy and mutual-help group treatment of conjugal bereavement. Am J Psychiatry, 145(2), 203-209. McDermut, W., Miller, I., & Brown, R. (2001). The efficacy of group psychotherapy for depression: A meta-analysis and review of the empirical research. Clinical Psychology: Science and Practice, 8(1), 98-116. McLean, P. D., & Hakstian, A. R. (1979). Clinical depression: Comparative efficacy of outpatient treatments. Journal of Consulting and Clinical Psychology, 47, 818-836. Meares, R., Stevenson, J., & Comerford, A. (1999). Psychotherapy with borderline patients: I. A comparison between treated and untreated cohorts. Australian-and-New-Zealand-Journalof-Psychiatry, 33(4), 467-472. Miller, W. R., & Hester, R. K. (1986). The effectiveness of alcoholism treatment: What research reveals. In W. R. Miller & N. Heather (Eds.), Treating addictive behaviors; the process of change. New York: Plenum Press. Miller, W. R., & Wilbourne, P. L. (2002). Mesa Grande: A methodological analysis of clinical trials of treatments for alcohol use disorders. Addiction, 97, 265-277. Miller, W. R., Wilbourne, P. L., & Hettema, J. E. (2003). What works? A summary of alcohol treatment outcome research. In R. K. Hester & W. R. Miller (Eds.), Handbook of Alcoholism Treatment Approaches: Effective Alternatives (3rd ed) (pp. 13-63). Boston, MA: Allyn & Bacon. Milrod, B., & Busch, F. (1996). Long-term outcome of panic disorder treatment. A review of the literature. J Nerv Ment Dis, 184(12), 723-730. Milrod, B., Busch, F., Cooper, A., & Shapiro, T. (1997). Manual of Panic - Focused Psychodynamic Psychotherapy. Washington, D.C.: .APA Press,. Milrod, B., Busch, F., Leon, A. C., Aronson, A., Roiphe, J., Rudden, M., et al. (2001). A pilot open trial of brief psychodynamic psychotherapy for panic disorder. J Psychother Pract Res, 10, 239-245. Milrod, B., Busch, F., Leon, A. C., Shapiro, T., Aronson, A., Roiphe, J., et al. (2000). Open trial of psychodynamic psychotherapy for panic disorder: a pilot study. Am J Psychiatry, 157(11), 1878-1880.

173

Milrod, B., Cooper, A., & Shear, M. (2001). Psychodynamic concepts of Anxiety. In D. Stein & E. Hollander (Eds.), Textbook of Anxiety Disorders (pp. 89-103). Washington, D.C.: American Psychiatric Press. Monsen, J. T., Odland, T., Faugli, A., Daae, E., & Eilertsen, D. E. (1995). Personality disorders: Changes and stability after intensive psychotherapy focusing on affect consciousness. Psychotherapy Research, 5, 33-48. Moran, G., Fonagy, P., Kurtz, A., Bolton, A., & Brook, C. (1991). A controlled study of psychoanalytic treatment of brittle diabetes [see comments]. Journal Of The American Academy Of Child And Adolescent Psychiatry, 30(6), 926-935. Mynors-Wallis, L., Davies, I., Gray, A., Barbour, F., & Gath, D. (1997). A randomised controlled trial and cost analysis of problem-solving treatment for emotional disorders given by community nurses in primary care. Br J Psychiatry, 170, 113-119. Mynors-Wallis, L. M., Gath, D. H., Day, A., & Baker, F. (2000). Randomised controlled trial of problem solving treatment, antidepressant medication, and combined treatment for major depression in primary care. BMJ, 320(7226), 26-30. Mynors-Wallis, L. M., Gath, D. H., Lloyd-Thomas, A. R., & Tomlinson, D. (1995). Randomised controlled trial comparing problem solving treatment with amitriptyline and placebo for major depression in primary care. BMJ, 310(6977), 441-445. Nordhus, I. H., & Pallesen, S. (2003). Psychological treatment of late-life anxiety: an empirical review. J Consult Clin Psychol, 71(4), 643-651. Ogrodniczuk, J. S., Piper, W. E., McCallum, M., Joyce, A. S., & Rosie, J. S. (2002). Interpersonal predictors of group therapy outcome for complicated grief. International Journal of Group Psychotherapy, 52, 511-535. Olfson, M. (1999). Emerging methods in mental health outcomes research. Journal of Practical Psychiatry and Behavioural Health, 5, 20-24. Olfson, M., Mechanic, D., Boyer, C. A., & Hansell, S. (1998). Linking inpatients with schizophrenia to outpatient care. Psychiatric Services, 49, 911-917. Palmer, C., & Fenner, J. (1999). Getting the message across: Review of research and theory about disseminating information within the NHS. London: Gaskell. Pattison, E. M., Brissenden, A., & Wohl, T. (1967). Assessing specific aspects of in-patient group psychotherapy. International Journal of Group Psychotherapy, 17, 283-297. Perl, M., Westlin, A. B., & Peterson, L. G. (1985). The female rape survivor: Time limited therapy with female/male co-therapists. Journal of Psychosomatic Obstetrics and Gynaecology, 4, 197-205. Perry, J. C., Banon, E., & Ianni, F. (1999). Effectiveness of psychotherapy for personality disorders. Am J Psychiatry, 156(9), 1312-1321. Piper, W. E. (1996). in #1716 Time-;imited. Washington. Piper, W. E., Azim, H. F. A., Joyce, A. S., McCallum, M., Nixon, G. W. H., & Segal, P. S. (1991). Quality of object relations vs interpersonal functioning as a predictor of the therapeutic alliance and psychotherapy outcome. Journal of Nervous and Mental Disease, 179, 432438. Piper, W. E., Joyce, A. S., McCallum, M., & Azim, H. F. (1998). Interpretive and supportive forms of psychotherapy and patient personality variables. J Consult Clin Psychol, 66(3), 558-567. Piper, W. E., McCallum, M., Joyce, A. S., Rosie, J. S., & Ogrodniczuk, J. (2001). Patient personality and time-limited group psychotherapy for complicated grief. International Journal of Group Psychotherapy, 51, 525-552.

174

Piper, W. E., Ogrodniczuk, J. S., McCallum, M., Joyce, A. S., & Rosie, J. S. (2003). Expression of affect as a mediator of the relationship between quality of object relations and group therapy outcome for patients with complicated grief. J Consult Clin Psychol, 71(4), 664671. Pomerleau, O., Pertshuk, M., Adkins, D., & d'Aquili, E. (1978). Treatment for middle-income problem drinkers. In P. E. Nathan, G. A. Marlatt & T. Loberg (Eds.), Alcoholism: New directions in behavioral research and treatment. New York: Plenum Press. Prendergast, M. L., Podus, D., Chang, E., & Urada, D. (2002). The effectiveness of drug abuse treatment: a meta-analysis of comparison group studies. Drug Alcohol Depend, 67(1), 5372. Radford, P., Wiseberg, S., & Yorke, C. S. B. (1972). A Study of "Main-Line" Heroin Addiction A Preliminary Report. Psychoanal. Study Child, 27, 156-180. Reiss, D., Grubin, D., & Meux, C. (1996). Young psychopaths in special hospital: treatment and outcome. British Journal of Psychiatry, 168, 99-104. Rockland, L. H. (1987). A supportive approach: Psychodynamically oriented supportive therapy treatment of borderline patients who self-mutilate. Journal of Personality Disorders, 1, 350-353. Roth, A., & Fonagy, P. (in press). What works for whom? A critical review of psychotherapy research. Second edition. New York: Guilford Press. Roth, S., Dye, E., & Lebowitz, L. (1988). Group therapy for sexual-assault victims. Psychotherapy, 25, 82-93. Rounsaville, B. J., Glazer, W., Wilber, C. H., Weissman, M. M., & Kleber, H. D. (1983). Shortterm interpersonal psychotherapy in methadone-maintained opiate addicts. Arch Gen Psychiatry, 40(6), 629-636. Russell, G. F. M., Szmukler, G. I., Dare, C., & Eisler, I. (1987). An evaluation of family therapy in anorexia nervosa and bulimia nervosa. Archives of General Psychiatry, 44, 1047-1056. Ryle, A. (1990). Cognitive Analytic Therapy: Active Participation in change. Chichester: Wiley. Ryle, A., & Golynkina, K. (2000). Effectiveness of time-limited cognitive analytic therapy of borderline personality disorder: factors associated with outcome. Br J Med Psychol, 73(Pt 2), 197-210. Sandahl, C., Herlittz, K., Ahlin, G., & Rnnberg, S. (1998). Time-limited group therapy for moderately alcohol dependent patients: a randomised controlled trial. Psychotherapy Research, 8, 361-378. Sandell, R., Blomberg, J., Lazar, A., Carlsson, J., Broberg, J., & Rand, H. (2000). Varieties of long-term outcome among patients in psychoanalysis and long-term psychotherapy: a review of findings in the Stockholm outcome of psychoanalysis and psychotherapy project (STOPP). International Journal of Psychoanalysis, 81(5), 921-943. Saunders, D. G. (1996). Feminist cognitive-behavioral and process-psychodynamic treatments for men who batter: interaction of abuser traits and treatment models. Violence and Victims, 11, 393-414. Scarvalone, P., Cloitre, M., & Difede, J. (1995). Interpersonal process therapy for incest survivors: preliminary outcome data. Paper presented at the Society for Psychotherapy Research Conference, Vancouver. Sensky, T., & Scott, J. (2002). All you need is cognitive behaviour therapy? Critical appraisal of evidence base must be understood and respected. BMJ, 324(7352), 1522.

175

Shapiro, D. A., Barkham, M., Rees, A., Hardy, G. E., Reynolds, S., & Startup, M. (1994). Effects of treatment duration and severity of depression on the effectiveness of cognitivebehavioral and psychodynamic-interpersonal psychotherapy. J Consult Clin Psychol, 62(3), 522-534. Shapiro, D. A., Rees, A., Barkham, M., Hardy, G., Reynolds, S., & Startup, M. (1995). Effects of treatment duration and severity of depression on the maintenance of gains following cognitive behavioral and psychodynamic interpersonal psychotherapy. Journal of Consulting and Clinical Psychology, 63, 378-387. Shea, M. T., Pilkonis, P. A., Beckham, E., Collins, J. F., Elkin, I., Sotsky, S. M., et al. (1992). Personality disorders and treatment outcome in the NIMH treatment of depression collaborative research program. American Journal of Psychiatry, 147, 711-718. Shear, M. K., Houck, P., Greeno, C., & Masters, S. (2001). Emotion-focused psychotherapy for patients with panic disorder. Am J Psychiatry, 158(12), 1993-1998. Shear, M. K., Pilkonis, P. A., Cloitre, M., & Leon, A. C. (1994). Cognitive behavioral treatment compared with non-prescriptive treatment of panic disorder. Archives of General Psychiatry, 51, 395-401. Shefler, G., Dasberg, H., & Ben-Shakhar, G. (1995). A randomised controlled outcome and follow-up study of Mann's time-limited psychotherapy. Journal of Consulting and Clinical Psychology, 63, 585-593. Simpson, D. D., Joe, G. W., & Broome, K. M. (2002). A national 5-year follow-up of treatment outcomes for cocaine dependence. Arch Gen Psychiatry, 59(6), 538-544. Simpson, D. D., Joe, G. W., Fletcher, B. W., Hubbard, R. L., & Anglin, M. D. (1999). A national evaluation of treatment outcomes for cocaine dependence. Arch Gen Psychiatry, 56(6), 507-514. Stanton, A. H., Gunderson, J. G., Knapp, P. H., Vancelli, M. L., Schnitzer, R., & Rosenthal, R. (1984). Effects of psychotherapy in schizophrenia: I. Design and implementation of a controlled study. Schizophrenia Bulletin, 10, 520-563. Steuer, J., Mintz, J., Hammen, C., Hill, M. A., Jarvik, L. F., McCarley, T., et al. (1984). Cognitivebehavioral and psychodynamic group psychotherapy in treatment of geriatric depression. Journal of Consulting and Clinical Psychology, 52, 180-192. Stevenson, J., & Meares, R. (1992). An outcome study of psychotherapy for patients with borderline personality disorder. American Journal of Psychiatry, 149, 358-362. Stevenson, J., & Meares, R. (1999). Psychotherapy with borderline patients: II. A preliminary cost benefit study. Australian-and-New-Zealand-Journal-of-Psychiatry, 33(4), 473-477. Svartberg, M., Stiles, T. C., & Seltzer, M. H. (2004). Randomized, controlled trial of the effectiveness of short-term dynamic psychotherapy and cognitive therapy for cluster C personality disorders. Am J Psychiatry, 161(5), 810-817. Swartz, M., Blazer, D., & Winfield, I. (1990). Estimating the prevalence of borderline personality disorder in the community. Journal of Personality Disorders, 4, 257. Target, M., & Fonagy, P. (1994a). The efficacy of psychoanalysis for children with emotional disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 361371. Target, M., & Fonagy, P. (1994b). The efficacy of psychoanalysis for children: Developmental considerations. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 1134-1144.

176

Tarrier, N. (2002). Commentary: Yes, cognitive behaviour therapy may well be all you need. British Medical Journal, 324, 288-294. Taylor, R. (2000). A seven year reconviction study of HMP Grendon Therapeutic Community. Home Office. Teasdale, J. D., Segal, Z. V., Williams, J. M., Ridgeway, V. A., Soulsby, J. M., & Lau, M. A. (2000). Prevention of relapse/recurrence in major depression by mindfulness- based cognitive therapy. J Consult Clin Psychol, 68(4), 615-623. Thase, M. E., Greenhouse, J. B., Frank, E., Reynolds, C. F., Pilkonis, P. A., Hurley, K., et al. (1997). Treatment of major depression with psychotherapy or psychotherapypharmacotherapy combinations. Archives of General Psychiatry, 54, 1009-1015. Thompson, L. W., Gallagher, D., & Breckenridge, J. S. (1987). Comparative effectiveness of psychotherapies for depressed elders. Journal of Consulting and Clinical Psychology, 55, 385-390. Thompson-Brenner, H., Glass, S., & Westen, D. (2003). A multidimensional meta-analysis of psychotherapy for bulimia nervosa. Clinical Psychology: Science and Practice, 10, 269287. Thornton, D., Mann , R., Bowers, L., & al. (1996). Sex offenders in a therapeutic community. In J. Shine (Ed.), Grendon: A compilation of Grendon research. Tomsovic, M. (1970). A follow-up study of discharged alcoholics. Hospital and Community Psychiatry, 21, 94-97. Torgersen, S., Kringlen, E., & Cramer, V. (2001). The prevalence of personality disorders in a community sample. Archives of General Psychiatry, 58, 590-596. Tucker, L., Bauer, S. F., Wagner, S., Harlam, D., & Sher, I. (1987). Long-term hospital treatment of borderline patients: A descriptive outcome study. American Journal of Psychiatry, 144, 1443-1448. Waldinger, R. J., & Gunderson, J. G. (1984). Completed therapies with borderline patients. American Journal of Psychotherapy, 38, 190-202. Wallerstein, R. S. (1986). Forty-two Lives in Treatment: A Study of Psychoanalysis and Psychotherapy. New York: Guilford Press. Wallerstein, R. S. (1989). The psychotherapy research project of the Menninger Foundation: an overview. Journal of Consulting and Clinical Psychology, 57, 195-205. Walsh, B. T., Wilson, G. T., Loeb, K. L., Devlin, M. J., Pike, K. M., Roose, S. P., et al. (1997). Medication and psychotherapy in the treatment of bulimia nervosa. Am J Psychiatry, 154(4), 523-531. Wampold, B. E. (1997). Methodological problems in identifying efficacious psychotherapies. Psychotherapy Research, 7, 21-43. Wampold, B.E., Minami, T., Baskin, T., & Callen-Tierney, S. (2002). A meta-(re)analysis of the effects of cognitive therapy versus 'other therapies' for depression. Journal of Affective Disorders, 68(2-3), 159-165. Warren, F., Preedy-Fayers, K., McGauley, G., Pickering, A., Norton, K., Geddes, J. R., et al. (2003). Review of treatments for severe personality disorder Home Office Online Report 30/03 (http://www.homeoffice.gov.uk/rds/pdfs2/rdsolr3003.pdf). Weisz, J. R., Weiss, B., Han, S. S., Granger, D. A., & Morton, T. (1995). Effects of psychotherapy with children and adolescents revisited: a meta-analysis of treatment outcome studies. Psychol Bull, 117(3), 450-468.

177

Weisz, J. R., Weiss, B., Morton, T., Granger, D., & Han, S. (1992). Meta-analysis of psychotherapy outcome research with children and adolescents. Los Angeles: University of California. Westen, D., & Morrison, K. (2001). A multidimensional meta-analysis of treatments for depression, panic, and generalized anxiety disorder: An empirical examination of the status of empirically supported therapies. Journal of Consulting and Clinical Psychology, 69, 875-899. Westen, D., Morrison, K., & Thompson-Brenner, H. (2004). The empirical status of empirically supportd psychotherapies: Assumptions, findings, and reporting in controlled clinical trials. Psychological Bulletin. Wetherell, J. L., Gatz, M., & Craske, M. G. (2003). Treatment of generalized anxiety disorder in older adults. J Consult Clin Psychol, 71(1), 31-40. Whittle, P. (2000). Experimental psychology and psychoanalysis: What we can learn from a century of misunderstanding. Neuro-psychoanalysis, 1, 233-245. Wiborg, I. M., & Dahl, A. A. (1996). Does brief dynamic psychotherapy reduce the relapse rate of panic disorder? Arch Gen Psychiatry, 53(8), 689-694. Wilberg, T., Friis, S., Karterud, S., Mehlum, L., Urnes, O., & Vaglum, P. (1998). Outpatient group psychotherapy: A valuable continuation treatment for patients with borderline personality disorder treated in a day hospital? A three year follow-up study. Nordic Journal of Psychiatry, 52, 213-221. Winston, A., Laikin, M., Pollack, J., Samstag, L. W., McCullough, L., & Muran, J. C. (1994). Short-term psychotherapy of personality disorders. American Journal of Psychiatry, 151, 190-194. Winston, A., Pollack, J., McCullough, L., Flegenheimer, W., Kestenbaum, R., & Trujillo, M. (1991). Brief psychotherapy of personality disorders. Journal of Nervous and Mental Disease, 179, 188-193. Woody, G. E., Luborsky, L., McLellan, A. T., O'Brien, C. P., Beck, A. T., Blaine, J., et al. (1983). Psychotherapy for opiate addicts. Does it help? Arch Gen Psychiatry, 40(6), 639-645. Woody, G. E., McLellan, A. T., Luborsky, L., & O'Brien, C. P. (1987). Twelve-month follow-up of psychotherapy for opiate dependence. American-Journal-of-Psychiatry, 144(5), 590596. Woody, G. E., McLellan, A. T., Luborsky, L., & O'Brien, C. P. (1990). Psychotherapy and counseling for methadone-maintained opiate addicts: Results of research studies. In L. S. Onken & J. D. Blaine (Eds.), Psychotherapy and Counseling in the Treatment of Drug Abuse. NIDA Research Monograph 104. Woody, G. E., McLellan, A. T., Luborsky, L., & O'Brien, C. P. (1995). Psychotherapy in community methadone programs: a validation study. Am J Psychiatry, 152(9), 1302-1308. Yalom, I. D. (1995). The Theory and Practice of Group Psychotherapy (4th ed.). New York: Basic Books. Young, J. E. (1999). Cognitive Therapy for Personality Disorder: A Schema-Focused Approach (3rd ed.). Sarasota, FL, US: Professional Resource Press/Professional Resource Exchange, In. Zimberg, S. (1974). Evaluation of alcoholism treatment in Harlem. Quarterly Journal of Studies on Alcohol, 35, 550-557.

178

Zimmerman, M., & Coryell, W. H. (1990). Diagnosing personality disorders within the community: A comparison of self-report and interview measures. Archives of General Psychiatry, 47, 527.

179